Full text of "Bulletin of the Museum of Comparative Zoology at Harvard College"

See other formats

Ylvl"-

HARVARD UNIVERSITY

Library of the

Museum of

Comparative Zoology

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LXIII. No. 1.

A SECTION IN THE TRENTON LIMESTONE AT
MARTINSBURG, NEW YORK.

By Thomas H. Clark.

With One Plate.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

May, 1919.

No. 1. — A Section in the Trenton Limestone at Marthisburg, New York.

By Thomas H. Clark.

Martinsburg is a small town situated in Lewis Co., N. Y., west
of the Adirondacks and east of Lake Ontario. It lies just within the
northwestern corner of the "Port Leyden quadrangle," the geology
of which has been described by Prof. W. J. Miller (N. Y. state mus.
Bull. 1910, 135). The village is situated upon the upper of two
prominent terraces facing the Black River, and at an elevation of
about 500 feet above it. The contact of the Pre-Cambrian gneiss
with the sediments is along the western margin of the alluvial plain
of the Black River, and the terraces are underlain by Ordovician
limestone which dips gently westward.

The summit floor of the terrace upon which Martinsburg stands
varies in width from one to two miles, and between it and the level
land along the Black River is a steep slope broken by a narrow but
pronounced shelf developed on the surface of the Black River lime-
stone. In this slope Roaring Brook has cut a deep trench, which
affords a fresh and almost continuous section of the Ordovician lime-
stone. The exposures along this brook make up the greater part
of the section here described, but as the rocks are largely concealed
along its upper stretches, outcrops along the road and old quarries
near the village of Martinsburg were also studied.

In the Bulletin mentioned above. Miller has given a detailed section
of the strata below the Trenton, but concerning that formation he
gives no information beyond the estimate of 175 feet for the total
thickness.

Dr. Raymond visited this section in 1912, and has called attention
(Summary report Director Geol. survey. Department of mines,
Canada, for 1912, 1914, p. 345) to its importance in the correlation
of the Trenton of the typical section at Trenton Falls, N. Y., with
the supposedly equivalent deposits in Ontario. His work indicated
the desirability of more detailed study of the section, and at his
request I spent two weeks in the area in 1916, and visited it again for
a short time in 1917.

THE SECTION.

For convenience, the beds of the section will be described in ascend-
ing order.

The hard cherty Black River limestone has resisted erosion more

4 bulletin: museum of comparative zoology.

successfully than the rather thinly bedded Trenton limestone above it,
so that a platform a quarter of a mile wide has been excavated upon
its surface. Since the surface of this platform, even in the stream
bed, is more or less masked by debris from the bluff behind it, the
exact contact with the Trenton is seldom seen. Fortunately, how-
ever, I found on my second visit that the spring freshets of 1917 had
removed a part of the covering of detritus in the stream bed, so that
the basal twelve inches of the Trenton was clearly shown. This
proved to be a dark blue, fine grained dense limestone, mostly without
fossils, but with here and there a thin band of small specimens. This
bed was without shaly partings to divide it into subordinate layers,
and showed no trace of clastic matter, so there was no evidence of a
"basal conglomerate." It differed from the underlying Black River
limestone in being finer grained with a more flinty fracture, and in
lacking the chert. The few fossils obtained from it are species char-
acteristic of the Trenton, Triplecia extans being the most important.
Above this basal layer there would seem to be about five feet of
strata concealed, the next outcrop being in the bank of the stream
about halfway across the shelf. At this locality a few thick beds of
highly fossiliferous dark blue impure limestone are exposed, with
rather thick bands of shale between the beds of limestone. The
following were the most common fossils, the first three being particu-
larly diagnostic, and, in this section, apparently confined to the lower
ten feet of the Trenton : —

Triplecia extans (Emmons), T. cuspidata (Hall), T. schucherti, sp.
nov., Parastrophia hcmiplicata Hall, Rafincsquina altcrnata (Emmons),
Suhulites elongahu Conrad, Honnoioma trentonensis Ulrich & Scofield,
H. bellicincta (Hall), Streptelasma corniculum Hall.

The last species was particularly abundant; one slab, about thirty-
six square inches in area, containing about forty specimens on its
surface. From the partings between the beds of limestone, large
numbers of complete specimens of Triplecia, retaining both valves,
were obtained.

Above the fossiliferous basal beds there are about 280 feet of dark
limestone in beds two inches to a foot in thickness, separated by
partings of shale which become more numerous and thicker in the
upper part of this division, but which seldom contain any great nuia-
ber of fossils. Some of the beds of limestone are composed of fine
grained material, while others contain a certain amount of more
coarsely crystalline calcite. In the latter case the crystalHzation
affects only the matrix, and does not obscure the fossils. Some of

CLARK: TRENTON LIMESTONE AT MARTINSBURG. O

the layers are made up almost entirely of fragments of shells, a sort
of coquina, while others are nearly or quite devoid of fossils.

The peculiarities of the distribution of fossils in these strata are
as follows : —

At seventy feet above the base of the Trenton there is a layer,
exposed on the road from Martinsburg to East Marti nsburg, which
contains great numbers of Sinuites cancellatus (Hall).

One hundred feet above the base there is a layer containing, among
other fossils, Dalmanclla rogata (Sardeson), Isotelus gigas DeKay,
Calymene senaria Conrad, Cryptolithus tcssellatus Green.

This layer is foimd near the base of the upper of the three falls
which have been developed by Roaring Brook in the lower part of
the Trenton, and five feet above the broad flat platform which extends
from the top of the middle fall to the base of the upper one. The
layer containing Crj^tolithus is only a few inches thick, and is a
fairly pure, nearly black limestone. Cryptolithus is quite abundant
and its discovery here is of considerable interest as it has not previously
been reported in the Trenton north of Trenton Falls in the region
west of the Adirondacks.

In the strata 100 feet above the base Platystrophia first becomes
abundant, and the fifteen feet above the 165 foot level contain espe-
cially well-preserved specimens. Associated with the Platystrophias
are the other common fossils of the Trenton, such as Calymene senaria,
Dalmanella rogata, Pledambonites sericeus, etc.

Between the 200 and 300 foot levels the limestone is composed of
httle else than fragments of shells, and at 280 feet is the lowest layer
of coarsely crystalUne limestone. Large specimens of an undescribed
Dalmanella are found in this stratum, and as usual, only the matrix
and not the fossils is affected by the crystallization. At 290 feet
there is a layer showing the so-called "giant ripple mark," the crests
being several feet apart, and their^ direction about N. 45° E. The
rock is composed of columnals of crinoids and fragments of brachio-
pods. In the hundred feet of beds just described there are no very
remarkable fossils, but it is the zone of the greatest development of
Prasopora simulatrix, which is there very common indeed.

Between 300 and 390 feet above the base the rocks are a rather
coarsely crystalhne limestone which is neither very well exposed nor,
apparently, very fossiliferous. In the basal layer I found a Carneyella,
the first representative of the Agelacrinitidae to be reported from the
Trenton of New York; and this layer is also the lowest bed in which
Rafinesquina deltoidea is found.

bulletin: museum of comparative zoology.

The remaining eighty-five feet of the Trenton, (390-475 ft.) is a
dark, thickly bedded, impure Umestone with little shale, which on
weathering breaks down into a rubbly mass. Fossils are rather
common in places, but only on weathered surfaces. Strophomena
trilobata, Rafincsquina camerata, and R. deltoidea are the most common
and characteristic fossils, while Hormotoma trentonensls, Trochonema
umbilicatuni, and Streptelasma corniculum are other abundant species,
these latter forms being "recurrent" from the lower twenty feet of
the formation.

The accompanying table shows the species found, and their vertical
distribution.

Si
CO

1—1

1
o

o
o

1— I

CD
I— I

00
<N

o
o

■*

a
o

a
o
u

Plantae

Buthotrenis caesoetosa Hall

Anthozoa

Streptelasma corniculum Hall

ECHINODERMATA

Cheirocrinus anatiformis(Hall)

X
X

Camevella ravmondi Clark

Bryozoa

Prasopora simulatrix var.
occidentalis Ulrich

X
X
X

Escharopora confluens Ulrich .
Rhindictya neglecta Ulrich. .

OSTRACODA

T/pnerditia sn. ind

X
X

Tkilobita

Cryptolithus tessellatus Green
Tsotelus ffiffas DeKav

X
X

Tllfleniis americanus Billinss

Calymene senaria Conrad ....
Ceraurus pleurexanthemus
Green .

X
X

Bkachiopoda

Pseudolingula rectilateralis . . .
^Emmons)

CLARK: TRENTON LIMESTONE AT MARTINSBURG.

1— 1

o
o

I— (

»-l

CD
I— t

§5

§
g

Trematis terminalis (Emmons) . .

Schizocrania filosa Hall

Plectorthis sp

Platystrophia sp. nov

"biforata"

colbyensis Foerste

Dalmanella rogata (Sardeson) . .
sp. nov

X
X
X

Dinorthis pectinella (Emmons)

X
X
X

X
X

X
X
X

X
X

Plectambonites sericeus (Sow-
erbv)

Rafinesquina alternata

(Emmons)

deltoidea (Conrad)

camerata (Conrad)

minnesotensis (N. H. Winchell)

X
X

Strophomena filitexta Hall

trilobata (Owen)

Triplecia cuspidata (Hall)

X
X
X

schucherti Clark

Parastrophia hemiplicata Hall . .

rotunda (Winchell & Schu-

Rhynchotrema increbescens
(Hall)

Zygospira recurvirostris (Hall) . .

X
X

Gastropoda

Sinuites cancellatus (Hall) . . .

X
X

V. corrugatus (Hall)

Phragmolites compressus Con.
Tetranota bidorsata (Hall) . . .

X
X

Pleurotomaria (Trochonema?)
ambigua Hall

Lophospira serrulata (Salter) .
Hormotoma trentonensis Ul-
rich & Scofield

X
X

bulletin: museum of comparative zoology.

i-H

I-H

J:
to

I-t

1
g

■*

Hormotoma bellicincta (Hall)

gracilis (Hall)

Liospira americana (Billings) .

Raphistoma rotuloides (Hall) .

Trochonema iimbilicatum
(Hall)

Gyronema percariDatum(Hall)

X
X
X

Holopea obKqiia Hall

X
X

Subulites elongatus Conrad . . .

Pelecypoda

Ctenodonta nasuta (Hall) ....

levata (Hall)

Whitella subtruncata (Hall) . .

Ambonychia amygdalina Hall

CORRELATION

Obviously this section should first be compared with that at Trenton
Falls, which is about sixty miles south of Martinsburg. Compar-
ing it with the composite Trenton Falls-Rathbone Brook section,
pubHshed by Ra;^Tnond, (Bull. M. C. Z., 1916, 56, p. 253), it appears
that the Trenton at Martinsburg is 129 feet thicker than at Trenton
Falls. The discovery of Cryptolithus at Martinsburg gives a much
better basis for comparison of the two sections than was previously
to be had. In the typical section, the highest layer with Cryptolithus
is seventy-three feet above the base, while at Martinsburg it is 100
feet above. At Trenton Falls, Rafinesquina deltoidea appears 228
feet above the base of the Trenton; at Martinsburg the lowest layer
containing this species is 300 feet above the base. In both sections
the strata between those characterized by these "guide fossils" are
layers of thinly bedded dark limestone alternating with beds of shale
one or two inches in thickness, the fauna in both cases being composed
of large numbers of the more common Trenton fossils, particularly
Prasopora simulatrix, Dalmanella, Plectambonites, Platystrophia,
Calymene, Ceraurus, and Isotelus. The similarity of the strata and

CLARK: TRENTON LIMESTONE AT MARTINSBURG. 9

the occurrence of these three zones in the same order are, I beheve,
sufficient to indicate the correctness of the correlation of the portions
of the sections occupied by these faunas, even though the zones of
the two localities are not of the same thickness. The lower 390 feet
of the section at Martinsburg are therefore probabl}^ the equivalent
of the total 346 feet of the combined Trenton Falls-Rathbone
Brook section, and the upper eighty-five feet of the section at Martins-
burg have no equivalent in the limestone of the more southern section,
but form a younger deposit. This is the view already advanced by
Raymond, but supported by less evidence.

While the general correlation between these two sections is obvious,
detailed correlation is attended with difficulties, and probably should
not be attempted until the intervening region can be studied. While
Cryptoliihus tessellatus occupies a zone about forty feet thick in which
it is abundant over a stretch of more than a thousand miles from east
of Quebec down past Montreal, the Champlain Valley, Saratoga,
and the Mohawk Valley to Rathbone Brook, at Martinsburg it is
found in only a single layer, and that twenty to thirty feet higher in
the section than it is normally found. Its western migration was
evidently delayed, and finally stopped by something other than a
physical barrier, but just what it was is not evident. At Martinsburg
the lowest beds are characterized especially by three species of Triple-
cia. At Trenton Falls one of these species, T. cxtans, is quite common
in one layer, but that layer is about seventy-five feet above the base,
and above tlie range of Cryptolithus, and not below it. The meaning
of this interchange of position is likewise not yet understood.

The occurrence of Triplecia in the lowest zone at Martinsburg is
of value in making a correlation with the Trenton of Ontario. In
the section at Ottawa and vicinity (Raymond, Guide book 3,
Excurs. 12th Internat. geol. cong. 1913) one finds at the base thirty-
five feet of limestone with T. extans, Phragmolites compressus, and
other fossils; in the middle a thick zone with numerous fossils in-
cluding an abundance of Prasopora simulatrix. Then come two
zones characterized by Rafinesquina deltoidea and Strophomena tri-
lobata respectively, the general sequence being that at Martinsbiu-g
but with a different development of the zone between that of Triplecia
and that of Prasopora.

These notes should suffice to show the great importance of the
Martinsburg section, since by combining in itself elements of the
faunas of the Trenton to the north and the Trenton to the south, it
permits a correlation which has formerly been in doubt. Perhaps

bulletin: museum of comparative zoology.

the most significant feature is the demonstration of the fact that the
strata at Martinsburg, containing the fauna with Strophomena trilo-
bata, gastropods, and Streptelasma, the so-called "Fusispira beds,"
are younger than any of the limestone in the section at Trenton Falls.
This has a far-reaching effect, for the Fusispira beds extend across
Ontario and appear as the upper part of the Prosser limestone in
Minnesota, Wisconsin, and Iowa.

description of new or interesting SPECIES.i

COELENTERATA.

Streptelasma corniculum Hall.
Streptoplasma corniculmn Hall, Pal. N. Y., 1847, 1, p. 69, pi. 25, fig. la-le.

This species is one of the commonest in the lowest beds, but does
not appear again until horizon 390 is reached. Here, and again at
400, 410, and 440 feet above the base it is fairly abundant. Although
the specimens from the upper part of the Trenton are generally larger
than those from the basal beds, I cannot determine any specific
dift'erence between them. By making transverse and longitudinal
sections of large individuals I satisfied myself that these show
in their younger stages the characteristics of Hall's species (now
abandoned) <S. parvula. The specimens from the upper part of the
Trenton are well preserved, while those from the basal beds are for
the most part badly crushed. It was in these beds that I found this
species in such great abundance. One slab of rotten shale, about
one quarter of a square foot in area had on its surface no fewer than
forty specimens.

Echinodermata.

Cheirocrinus anatiformis (Hall).

Plate 1, fig. 17.

Echino-encrinites anatiformis Hall, Pal. N. Y. 1847, 1, p. 89, 318, pi. 29, fig.
4a^f.

The discovery of a few specimens of this rare species is of interest,
since the exact horizon of the fossil has not previously been known.
Hall stated that his specimens were from the "midst of the fossili-

1 Bull. 92, U. S. N. M. contains an extensive bibliography of the various species here
discussed.

CLARK: TRENTON LIMESTONE AT MARTINSBURG.

ferous portion of the Trenton limestone" at Turin, Lewis Co., N. Y.
I found a single plate, not absolutely identifiable, in the zone with
Cryptolithus, 100 feet above the base of the section, but better and
readily recognizable specimens were found in the Upper Trenton,
from 390 to 410 feet above the base. It seems probable that this is
the true horizon of the species. A species of Cheirocrinus, named
C. walcotti by Jaekel (Stammesgsch. Pelmat., 1899, p. 221, pi. 11,
fig. 8) is very common in certain layers low in the upper third of the
Trenton at Trenton Falls and, therefore, at a somewhat lower horizon
than the Cheirocrinus at Martinsburg. The name was proposed
by Jaekel largely because C. anatiformis was so poorly described as to
be almost unrecognizable, and a comparison of specimens shows that
there is very little if any difference between the two species. The
specimens from Martinsburg show the numerous pectinirhombs
which were ignored by Hall in both illustration and description, and
one which retains the plates of the upper part of the calyx shows a
series of small plates covering the ventral grooves. That Jaekel was
right in referring this species to Cheirocrinus instead of Echino-
encrinites is obvious.

Carneyella raymondi, sp. nov.

Plate 1, fig. 18, 19.

Specimen small, nearly circular in outline, with a peripheral ring
one fourth of the diameter in width. There are six rays, five of which
are straight, but ray II is bifurcated about one third the distance
from the center to the peripheral ring, and the anterior branch curved
in a contrasolar direction. The supraoral plates are damaged, and
only three can be seen, but these are large, fully three times as large
as the lateral covering plates. Their outlines are mostly obhterated.
The interradii are covered with relatively large imbricating plates,
but unfortunately the anal interradius is so poorly preserved as to
obscure the anal pyramid. None of the rays shows any trace of
auxiliary covering plates. The type (M. C. Z. 3,978) and only known
specimen is 8 mm. in diameter. It was found by the writer, attached
to a shell of Rafinesquina alternata, in a layer 300 feet above the base
of the Trenton in the gorge of Roaring Brook, near Martinsburg,
N. Y. The horizon is at the base of the Rafinesquina deltoidea zone.
So far as is known, this is the first agelacrinitid to be found in the
Ordovician rocks of New York State.

bulletin: museum of comparative zoology.

The most closely allied species is probably C. muUibrachiatus
(Raymond) .1 The type of that species has eight rays, though Dr.
Raymond suggests that the normal number may prove to be seven.
In the Canadian species, however, the periphery is still broader than
in C. raymondi, the width being equal to one third of the diameter.

Bryozoa.

Prasopora simulatrix var. occidentalis Ulrich.

Prasopora simulatrix var. occidentalis Ulrich, Pal. Minn., 1893, 3 p. 246 pi
16, fig. 1, 2, 6, 7.

•

This species, and possibly others of like form, is extremely abundant
from 160 to 280 feet above the base of the Trenton. It also occurs
practically throughout the lower part of the Trenton. I do not wish
to give the impression that all the hemispheric Bryozoa collected at
Martinsburg belong to this species. Such is probably not the case.
But from the scores of specimens collected, I examined sections of six
taken at random. All of these proved to be Prasopora simulatrix
var. • occidentalis. While other genera may be represented in the
many unexamined forms, the probability is that few, if any, do not
belong to Prasopora.

ESCHAROPORA CONFLUENS Ulrich.
Escharopora confluens Ulrich, Pal. Minn., 1893, 3, p. 171, pi. 13, fig. 1-11.

I found this bryozoan very abundant in the basal beds. It occurs
mostly as inch-long fragments not showing the mode of growth, but
by removing a large slab of limestone from the bed of the creek, I
imcovered some large branching specimens embedded in the soft
shaly parting. One specimen was found to branch three times in a
length of two inches.

Brachiopoda.

PSEUDOLINGULA RECTILATERALIS (Emmons).
Lingula rectilateralis Emmons, Geol. N. Y. Rept. 2d dist., 1842, p. 399, fig. 6.

This species occurred in Upper, Middle, and Lower Trenton. At
horizon 180 I found a specimen protruding downwards from an over-

J Ottawa naturalist, 24, p. 60, pi. 1, fig. 2.

CLARK: TRENTON LIMESTONE AT MARTINSBURG.

hanging stratum of limestone, the anterior margin being embedded
in the matrix, and the posterior margin free below the rock. This was
probably the position in which the animal lived, besides being the
position in which it died, standing erect in the mud, as do the Lingulae
to-day.

Plectorthis sp.

In the Upper Trenton, near the state road, I found one specimen of
Plectorthis, but, although it is in an excellent state of preservation
and almost complete, it could not be identified with any described
species. In a genus usually so common, I hesitate to make a new
species from a single specimen.

Rafinesquina minnesotensis (Winchell).

Strophomena minnesotensis Winchell, Ninth ann. rept. Geol. nat. hist, survey
Minn., 1881, p. 120.

This species is rather frequently listed of late, usually from forma-
tions of Black River or Stones River age. The horizon of the original
specimens was, however. Lower Trenton, and it is gratifying that its
first record from New York should be at that horizon. The fossil is
very difficult to identify unless one has the interior of the brachial
valve. Such a specimen was found, showing the large muscular
area, the divergent ridges, and the shallow depression circumscribing
the scars. It is also thickly dotted with "ovarian" markings.

Strophomena trentonensis Winchell & Schuchert.

Strophomena trentonensis Winchell & Schuchert, Pal. Minn,, 1893, 3, pt. 2,
p. 389, pi. 30, fig. 41.

While this species has a wide distribution, specimens are seldom
found. A pedicle valve obtained fifty feet above the base of the
Trenton has the muscle scars somewhat smaller than in the Minne-
sotan type, and the two divergent ridges which should appear beside
the median elevation in the muscular area are lacking. In spite of
these variations, the smooth interior of the pedicle valve leaves little
doubt of the correctness of the identification.

14 bulletin: museum of comparative zoology.

Triplegia cuspidata (Hall).

Plate 1, fig. 1-6.

Atrypa cuspidata Hall, Pal. N. Y., 1847, 1, p. 138, 318, pi. 33* (supplementary

plate), fig. la-h: Hall & Clarke, Pal. N. Y., 1892, 8, pt. 1, p. 270.
Triplecia extans Hall & Clarke, Loc. cit, 1892, pi. lie, fig. 1-3.

Without exception, Triplecia cuspidata is the most abundant fossil
in the basal ten feet of the Trenton at Martinsburg, and a large num-
ber of complete specimens retaining both valves were collected. Most
are large, specimens 18 mm. long and 25 mm. wide being common.
Smaller specimens are also present in some numbers, but not the very
young. Two specimens, one 9 mm. by 11 mm., another 20 mm. by 28
mm., probably represent the extremes of the range in size. The
smaller of these is 6 mm. thick, the larger 17 mm. It proves in
practice somewhat difficult to separate the various species of Triplecia.
It will be remembered that Hall, at the end of his first work on
Trenton fossils, came to the conclusion that Atnjpa extans and Atrypa
cuspidata were identical. Triplecia nucha is easily recognized by its
small size, sharp, narrow fold and sinus, and absence of radial striae.
Triplecia extans and T. cuspidata are both striate, the latter much more
strikingly so than the former. After examining a large number of
specimens from Watertown, Martinsburg, Trenton Falls, and else-
where, it would appear that T. extans is characterized by a rounded
sinus, T. cuspidata by an angular or grooved sinus, and a new species,
now to be named, by a flat-bottomed sinus.

In my judgment, one of the specimens figured by Hall and Clark
as Trijilecia extans, is a very typical T. cuspidata, and I have so as-
signed it above. It displays the typical angular sinus. Figures 6
and 7 of the same plate show the rounded sinus of T. extans.

The exact horizon at which the various species of Triplecia occur
has not yet been determined. Hall stated that T. cuspidata was
known to him as occurring only in the central part of the Trenton
limestone at Lowville, a town only five miles north of Martinsburg.
I^did not have time to search for the original locality, but from the
nearness of Lowville to Martinsburg it seems probable that the
original specimens were really from the base of the Trenton, which is
the only horizon in which this species is found at Martinsburg. In
the M. C. Z. there are a great many specimens of this species from
Watertown, N. Y., but unfortunately without exact data as to the
horizon in the Trenton from which they were obtained. There are

CLARK: TRENTON LIMESTONE AT MARTINSBURG. 15

also two large specimens in a fragment of rock from Trenton Falls.
The matrix has on it the impression of a specimen of Phragmolites,
thus showing that these two forms occur together at Trenton Falls
as well as at Mardnsburg.

Triplegia schucherti, sp. nov.
Plate 1, fig. 7-12.

Shell large, of about the same size and shape as that of T. cuspidata,
but generally a little plimiper. The length and thickness are nearly
equal, the thickness equalling the length in some cases. The beaks
of the opposite valves are so closely opposed as to practically eliminate
a cardinal area. The brachial valve has a relatively narrow rounded
fold which does not project at the front, and which is bordered by
very shallow depressions. The fold is not so high and angular as in
either T. cuspidata or T. nucha, but more like that of T. extans. The
sinus in the pedicle valve is rounded at the bottom in the posterior
part of its course, but flat at the front and not deep. While this shell
attains the large size of T. cuspidata, it evidently resembled T.
extans in its younger stages, and is probably most nearly aUied to
that species.

The surface is marked by both concentric and radial lines, the latter,
like those on T. exians, being most prominent toward the anterior
margin of the shell. Large specimens of both this species and T.
cuspidata often show obscure radial plications as well as striae.

The type (M. C. Z. 8,534) selected as the best preserved of a dozen
specimens at hand, is 22 mm. long, 28 mm. wide, and 21 mm. thick.
A smaller specimen is 17 mm. long, 22 mm. wide, and 15.5 mm.
thick. The largest is 24 mm. long, 31 mm. wide, and 20 mm. thick.

Seven of the specimens were collected by the writer from the base
of ]the Trenton at Martinsburg, and there are five more in the M. C. Z.
from the Trenton at Watertown, N. Y.

Parastrophia rotunda (Winchell & Schuchert).

Anasirophia? hemiplicata var. rotunda Winchell & Schuchert, Pal. Minn.,
1893, 3, p. 383, pi. 30, fig. 32-35.

While it is the common custom to refer all the forms of Parastrophia
found in the Trenton to one species, Winchell and Schuchert con-
ferred a varietal name on a very rotund form which they obtained
from the Galena (Prosser) limestone of Minnesota. I found in the
lowest Trenton at Martinsburg three specimens of an equally rotund

bulletin: museum of comparative zoology.

Parastrophia, which, differing in many respects from the original
specimens, may perhaps, considering the variabiHty of the members
of this genus, be included in the Minnesotan group, as a species rather
than a variety.

Winchell and Schuchert state that their variety is "distinguished
in having the length and width nearly equal, the valves more convex,
and the plications somewhat more pronounced in the fold and sinus
and less numerous in the lateral portions of the shell." No further
description is given, but the single specimen figured is large (15 mm.
long), has three broad plications on the fold, two in the sinus, and a
pair on either side of the fold and sinus.

The three specimens found at Martinsburg are fully as convex as
the Minnesotan specimen, but the largest is only 10 mm. long, and
all have more and narrower plications in the fold and sinus. Two of
them have four plications on the fold and three in the sinus, and the
third, the largest, has four in the sinus. Two of them have two and
one of them three plications on each side of the fold and sinus.

This rather detailed description is given because of the increasing
necessity of more definite knowledge of the limits of variation, and
the possibility of distinguishing species of Parastrophia.

Parastrophia hemiplicata Hall.

Airypa hemiplicata Hall, Pal. N. Y., 1847, 1, p. 144, pi. 33, fig. 10.

To assist in a study of Parastrophia hemiplicata which must ulti-
mately be made, it is worth while to record the characteristics of any
specimens whose horizon is definitely known. Fourteen specimens
were found in the lower thirty feet of the Trenton, but a number of
these were too poorly preserved to yield any satisfactory information.
The following table shows the principal characteristics:

Length

Width

Thickness

Plications on fold

In sinus

On one side

9 mm.

1.5

CLARK: TRENTON LIMESTONE AT MARTINSBURG.

It will be noted that all the specimens show plications on the sides
of the fold and sinus. Moreover, there are obviously two distinct
groups, one with four plications in the sinus of the young, and the
other with only three plications in the sinus of the adult. The first
five belong to the first group, and also, probably, the last one, while the
other four belong to the second. The two specimens at the head of
the list are evidently immature, as is shown by their thinness and their
very short plications. With further growth new plications are added
in the fold and sinus as well as at the sides, and number five is probably
a typical adult.

No very young specimen of the second group has been found, but
the presence of only three plications in the adult shows that it could
not have been developed from such young as numbers one and two.

Only one specimen was collected from the Upper Trenton, a pedicle
valve found about 400 feet above the base. It agrees with the second
type in the table above, being 10 mm. long, 12 mm. wide, with three
plications in the sinus and two on the sides.

Miss Wilson, in her studies of Parastrophia hcmiplicata from about
Ottawa (Mus. bull. 2, Geol. surv. Canada, 1914) found specimens
with three to five plications in the sinus and plications on the sides of
the fold in both Middle and Upper Trenton.

Gastropoda.

SiNUITES CANCELLATUS (Hall).

Bellerophon bilobatus Emmons, Geol. N. Y., 1842, 2, p. 392, fig. 6.

This species was found to be exceedingly common at horizon 70.
It also occurred at horizons 15, 20, 390, and 410, but was not seen in
the Middle Trenton. In the basal beds, where some other forms
were so common, it was absent. At horizon 70, in actual numbers,
as the following list shows, it many times exceeded in abundance all
other forms collected. The fossils from this horizon, with the number
of specimens collected, which the writer can attest to be a fair index
to their relative abundance, are listed below: —

9 Prasopora simulafrix v. occidentalis Ulrich. 1 Schizocrania filosa
Hall. 106 Sinuites canccllaivs (Hall). 6 S. cancellatus v. corrugalus
(Hall). 1 Plevrotomaria (Trochonema f) ombigua HaW. 1 Hormotoma
trenionensis (Ulrich &Scofield). 4 Ctenodonta levaia (Hall). 3 Caly-
mene senaria Conrad — a total of 131.

18 bulletin: museum of comparative zoology.

Crustacea.

Cryptolithus tessellatus Green.

Cryptolithus tessellatus Green, Monog. N. Amer. trilobites, 1832, p. 73, pi. 1,
fig. 4.

In Quebec, eastern New York, and central Pennsylvania, Crypto-
lithus tessellatus is one of the most common fossils in the lower part
of the Trenton, but in Ontario, Wisconsin, Minnesota, and Iowa it
has not yet been found. Even at Trenton Falls the species is so rare
that it has been found only a few times by local collectors, who form-
erly spent a great deal of time searching for rare fossils in that vicinity.
North of Trenton Falls and west of the Adirondacks it has not previ-
ously been reported.

The specimens, which were all found in one layer about 100 feet
above the base of the section, seem to be quite typical, though all
rather small. In front of the girder there are two rows of pits, and
back of it at the sides there are three rows bordering the bases of the
lateral mounds. In front of the glabella there are three rows of pits.
Eye-lines are absent as is usual in the adult.

EXPLANATION OF THE PLATE.

Clark. — Trenton Limestone at Martinsburg.

EXPLANATION OF THE PLATE.

Fig. 1-3. TripJecia cuspidata (Hall). Posterior, lateral, and anterior views
of a specimen of average size from the base of the Trenton at
Martinsburg. Natural size.

Fig. 4-6. The same species, from the same locality. Anterior views of
three specimens to show variation in shape of sinus, with per-
sistence of the central groove. Natural size.

Fig. 7-9. Triplecia schucherii Clark. Posterior, lateral, and anterior views
of the holotype, a large specimen from the base of Trenton at
Martinsburg. Natural size.

Fig. 10, 11. The same species, from the same locality. Posterior and ante-
rior views of a specimen with a very high narrow fold, but a
flat-bottomed sinus. Natural size.

Fig. 12. The same species from the same locality. A smaller specimen
with a low rounded fold, and a flat-bottomed sinus. Natural
size.

Fig. 13-15. Triplecia extans (Emmons) . Posterior, lateral, and anterior views
to show the high fold with the evenly rounded sinus. Natural
size.

Fig. 16. Phragmclites compressus Conrad. A photograph of a specimen from
the base of the Trenton at Martinsburg. Natural size.

Fig. 17. Cheirocrinus anatiformis (Hall). A photograph of two fragmentary
specimens from the upper part of the Trenton at Martinsburg,
to show the pectinirhombs. Natural size.

Fig. 18. Carneyella raymondi Clark. A photograph of the type, resting
on the anterior part of the pedicle valve of a Rafinesquina.
From the base of the Upper Trenton at Martinsburg. Twice
natural size.

Fig. 19. The same specimen. An enlarged drawing. 4.5 times natural size.

BULL. MUS.COMP. ZOOL.

Clark. Trenton Limestone. Plate 1

is3

.»#

7c<?

Bulletin of the Museum of Comparative ZoSlogy
AT HARVARD COLLEGE.
Vol. LXIII. No. 2.

SOME CRITICAL NOTES ON BIRDS.

By Outram Bangs and Thomas E. Penard.

CAMBRIDGE, MASS., U. S. A.:
PRINTED FOR THE MUSEUM.
June, 1919.

No. 2. — Some Critical Notes on Birds.
By Outram Bangs and Thomas E. Penard.

For some time past we have been working together on the Lafres-
naye Collection of birds which a few years ago the Boston Society
of Natural History most generously and wisely gave to the Museum
of Comparative Zoology.

We intend later to publish an account of the collection, its types,
a sketch of Lafresnaye's life, a list of his published papers, etc., etc.
Of this we have already done much.

In the meantime we publish the following rather random notes —
mostly changes in names and descriptions of new forms, noticed
while identifying the Lafresnaye types.

We are much indebted to Dr. C. W. Richmond for his valuable
opinion, always cheerfully given, on many points of involved nomen-
clature in connection with our study of the Lafresnaye Collection,
and to the authorities of the U. S. National Museum, the American
Museum of Natural History, and the Field Museum of Natural
History for the loan of specimens.

Pterodroma haesitata (Kuhl).

Procellaria haesitata Kuhl, Beitr. zooL, 1820, p. 142 ("Mers de

rinde").

Procellaria diaboUca Lafresnaye, Rev. zooL, 1844, p. 168 (Guade-
loupe).

Coiype. — M. C. Z. 73,221, Lafr. coll. 8,000.

Cotypc.—M. C. Z. 73,222, Lafr. coll. 8,002.

Procellaria vieridionalis Lawrence, Ann. Lye. nat. hist. N. Y., 1848,
4, p. 475.

There were originally three cotypes in the Lafresnaye Collection.
One of these, 8,001, was exchanged in 1886 with Prof. Alfred Newton.
The original labels of all three specimens are now missing.

Noble (Bull. M. C. Z., 1916, 60, p. 370) discusses this species at
length, and gives measurements of A. diaholica and of some specimens
which he considers different, and to which he applies the name Aestre-
lata haesitata (Kuhl). He assumes quite correctly that the larger

22 bulletin: museum of comparative zoology.

bird is Lafresnaye's P. diabolica. Computing Kuhl's meftsurements
of P. haedtata on the basis of the "Pied du Roi" and the "Frankfurt
A. M. inch," and assuming thatKuhl, as a student of Temminck, used
the French system of measurements, he comes to the conclusion that
Kuhl's P. hacsitata applies to the smaller bird and accordingly re-
stricts the name in that sense.

Fortunately Kuhl's type is available to-day in the Leyden Museum.
It was listed with another specimen by Schlegel (Mus. Pays-Bas.
Mon. Procellariae, 1863, p. 13) imder the name Procellaria haesitata
Kuhl. Schlegel gives measurements of both specimens, which indicate
a large bird.

Dr. E. D. van Oort of the Leyden Museum, at our request, has very
kindly examined Kuhl's specimens, and whites us as follows: —

"There are in the collection of our Museum two specimens of
Aestrdata hacsitata (Kuhl), both originally labelled by Temminck:
Procellaria hasitaia, Mers de I'lnde. They are both the types of
Procellaria hacsitata Kuhl (Beitr. 1820, p. 142) and of Procellaria hasi-
tata in Temminck's PI. col. V, 1826, pi. 416. Temminck's plate,
however, is not correct, the upper tail coverts, as well as the under
tail coverts, being pure white, not grey. The two specimens are
exactly alike. Schlegel mentions these in his catalogue of the Pro-
cellariae (Mus. Pays-Bas, Proc, Procellaria hacsitata, p. 14) but he
wrongly designates only No. 1 as the type of the species, and changes
Temminck's locahty "Mers de I'lnde" into "Ocean."

"The measurements of the two specimens are: —

Schlegel's Cat. No. 1 No. 2.

mm.

Wing (without pressing down)

310

Tail (from base of free tail feathers)

130

131

Bill (to angle of mouth)

Culmen

Tarsus

Middle toe with claw

55.5

" In my opinion haesitata and diabolica are the same."
It is ev-ident from the above that Kuhl's cotypes are both large
birds agreeing very well in size Avith the types of P. diabolica, except
that the wing measurement is even greater. We agree with Dr.
van Oort that P. haesitata Kuhl and P. diabolica Lafresnaye are the
same. Obviously the name haesitata should be used for the species.

BANGS AND PENARD: SOME CRITICAL NOTES ON BIRDS.

The two small specimens which Noble (loc. cit, p. 372) considered
a different species, to which he applied the name Aestrelata haesitaia,
we do not regard specifically different from the larger birds. The
difference in the nostril tubes pointed out by Noble does not appear
to be due to a normal concUtion, but rather one of distortion in drying.
Noble refers P. meridionalis Lawrence, which is a very large bird, to
his P. haentata on the ground that the nosti'il tubes (and coloration)
are like those of the small birds. In our opinion P. meridionalis is
P. haesitaia Kuhl with normal nostril tubes. At any rate there is
great variation in the specimens, and we refer the small birds which
Noble considered a different species to P. haesitaia Kuhl, rather than
to rename them on the e\adence at hand.

A female taken at Pittsfield, N. H., August, 1893, by H. W. Osgood,
now in the collection of William Brewster (46,076) has normal nostril
tubes, but is a small bird; it affords the following measurements: —
wing, 265; tail, 128; bill (to angle of mouth), 37; culmen, 32; tarsus,
35; middle toe, with claw, 50.

Lafresnaye's reference to two species, a larger and smaller, which
arrive at different times of the year and nest at different heights, is
interesting, but there is no reason for assuming that the two small
specimens in the M. C. Z. represent the smaller species referred to by
Lafresnaye. It is also interesting to note in this connection that
Pere Labat's illustration of the Diablotin (Nouv. voy., 1724, 2, p. 349)
represents a uniformly dark-colored bird, and on the next page he
writes, "son plumage est noir."

Herpetotheres cachinnans queribundus, subsp. nov.

Tijve.— M. C. Z. 7,792. Brazil: Pernambuco. J. C. Fletcher.

Characters. — Similar to H. cachinnans cachinnans (Linne) of Guiana
but paler, the upper parts browner, much less blackish, the lower parts
whiter; similar also to H. c. chapmaili Bangs and Penard of Mexico
but slightly smaller and with very much less spotting on axillars and
lining of wing.

Measurements.

Culmen

from

No.

Locality

Sex

Wing

Tail

Tarsus

Cere

7,792 M. C. Z. .

Pernambuco

258

196

23.5

110,476 U. S.N. M.

Santarem

257

198

16, .526

Parana

267

195

24 bulletin: museum of comparative zoology.

Remarks. — The new form is decidedly paler than H. c. cachvmians;
in this respect it is like the northern H. c. chapmani, from which it
differs in the almost immaculate hning of the wing. H. c. chapmani
has the axillars and under wing coverts much, often heavily, spotted
with rusty.

There are now five forms of this species, which, though close, are
recognizable; they are:^ —

Herpdoihcrcs cachintums cachinnans (Linne). Gviianas; Venezuela?

Herpetothcrcs cachinnans queribimdus Bangs and Penard. Southern
Brazil; Paraguay; Bolivia.

Hcrpetoiheres cachinnans maestus Bangs and Noble. Northwest
Peru; southwest Ecuador?

Herpefoihcres cachinnans fulvcsccns Chapman. Tropical zone of
Pacific coast from eastern Panama to Ecuador.

Hcrpetoiheres cachinnans chapmani Bangs and Penard. Mexico
to western Panama.

EupsiTTULA astec vicinalis, subsp. nov.

r?/^T.— M. C. Z. 48,482. Adult d^. Mexico: Tamaulipas. Alta-
mira. 24 December, 1908. E. B. Armstrong.

Characters.- — Similar to E. astcc astec (Souancc), Vera Cruz to Costa
Rica, but paler and brighter green above and much paler and greener,
less brownish, below; upper parts between Parrot-green and Calliste-
green (of Ridgway); throat and chest light yellowish olive; middle
of breast and belly Javel -green.

Measurements. — Type, adult cf ; wing, 141; tail, 119; culmen
from cere, 19. Topotype, M. C. Z. 48,480, adult 9 ; wing, 135; tail,
111; culmen from cere, 19.

Remarks. — While collecting for Dr. John C. Phillips in Tamaulipas,
F. B. Armstrong took ten specimens of this new paroquet at Altamira
in December, 1908. There is no variation in the series, and when laid
out beside a series of true E. astcc the Tamaulipas birds look almost
as if they represented a distinct species, so much paler and brighter
are they in color.

Pterophanes cyanopterus (Eraser).

Trochilvs cyanopterus Eraser ex. Loddiges's Ms., Proc. Zool. soc,
1840, p. 17 (no type-locality given).

Ornismya temminckii (not Ornismya temminckii Lesson, 1829 =
Heliomaster sqnamosns (Temminck)) Boissonneau, Rev. zool., 1839,
p. 354 (Bogota).

BANGS AND PENARD : SOME CRITICAL NOTES ON BIRDS. 25

The name which has so universally been used for Temminck's
Sapphire-wing unfortunately cannot prevail on account of preoccupa-
tion, and Fraser's cyanoptenis appears to be the first available for it.

Synallaxis brachyurus brachyurus Lafresnaye.

Synnalaxis hrachyurus Lafr., Rev. zooL, 1843, p. 290 (Colombia).

Type.— M. C. Z. 77,256, Lafr. coll. 2,456.

SynaUaxis pudica Sclater, Proc. Zool. soc. London, 1859, p. 191
(Bogota).

We have carefully coiiipared the type, an immature bird with a
short, undeveloped tail, with the various forms of Synallaxis pudica
and with SynaUaxis svbpndica Sclater. We find Synallaxis pudica
Sclater identical with <S. hrachyunis hrachyurus Lafr.

The subspecies now become : —

Synallaxis brachyurus brachy^irtis Lafresnaye.

Synallaxis brachyurus nigrifumosa Lawrence.

Synallaxis brachyurus cailcae Chapman and

Synallaxis brachyurus chapmani, subsp. nov.

Type.— M.C. Z. 124,478. Adult cf. Colombia: Jiminez,
tropical zone of Pacific slope. 6 April, 1904. M. G. Palmer.

Characters. — Most nearly hke S. h. nigrifumosa, but back less oliva-
ceous, more gra^,nsh, and under parts much paler and grayer. Chest-
nut of pileum and wings darker than in S. b. brachyurus and very much
darker than in S. b. caucae. Size large. Type, adult & ; wing, 66;
tail, 80; tarsus, 23; exposed culmen, 15.

Chapman (Bull. Amer. mus. nat. hist., 1917, 36, p. 404) has already
called attention to this race from western Colombia, but has not
named it. We consider it quite as well marked as any of the other
subspecies.

Dendrocincla lafresnayei christiani, subsp. nov.

Type.— M. C. Z. 124,522. Adult 9. Colombia: near Pavas^
Pacific slope of western Andes. 10 March, 1908. M. G. Palmer.

Characters. — Similar to D. I. lafresnayei Ridgway of eastern Colom-
bia, and of about the same size, but much darker in color, and more
ohvaceous, less brownish throughout; darker and more oKvaceous
even than S. I. ridgwayi Oberholser of Panama and Costa. Rica.

26 bulletin: museum of comparative zoology.

Measurements.— Type, adult 9; wing, 101; tail, 80; tarsus, 23;
exposed culmen, 26. Topotype, M. C. Z. 124,523, adult & ; wdng, 109;
tail, 88; tarsus, 23; exposed culmen, 27.

Remarks. — We have named this bird in honor of Christian de La-
fresnaye, Lt. Colonel d'artillerie en retraite, to whom we are indebted
for much assistance in bringing together material for a sketch of the
life of his illustrious father.

PiCOLAPTES AFFINIS LIGNICIDA, Subsp. nOV.

Type.— M. C. Z. 49,359. Adult d". Mexico: Tamaulipas, Ga-
lindo. 24 March, 1909. F. B. Armstrong.

Characters. — Similar to P. affinis affinis (Lafr.), but very much paler
and grayer throughout; back more olivaceous, less reddish brown;
head much grayer, less bro\\Tiish; under parts paler, grayer, and
wholly lacking the rich brown (raw umber) coloring of the under parts
in C. affinis affinis. Size about the same.

Measurements.— Type, adult &; wing, 108; tail, 95; tarsus, 21;
exposed culmen, 28. Topotypc, M. C. Z. 49,354, adult 9 ; wing, 107;
tail, 99; tarsus, 21.5; exposed culmen, 28.

Remarks. — Mr. Armstrong collected a series of eleven skins of this
strongly marked northern race of P. affinis among the very arid
tropical hills of the region north and west of Ciudad Victoria. At the
time Dr. J. C. Phillips reported on the Armstrong Collection (x\uk,
January, 1911, 28, p. 67) we lacked adequate material of true P. affinis,
though we had large series of P. affinis ncgledus Ridgway, and Dr.
Phillips considered the Armstrong skins to represent the northern
form as distinguished from neglectus. In reality the Tamaulipas
bird is very distinct, while affinis and neglectus are exceedingly close
forms, hardly distinguishable except in long series.

Muscisaxicola macloviana mentalis d'Orbigny and Lafresnaye.

Muscisaxicola mentalis d'Orb. and Lafr., Mag. zool. Syn. avium,
1837, p. 66 ("Habit. Cobija, in Boli\qa; Arica, atque Patagonia").

Cotype (?). — M. C. Z. 77,316, Lafr. coll. 4,599.

Muscisaxicola albimentum Lafresnaye, Rev. et mag. zool., 1855,
p. 61 (type-locality not stated, label gives "Bolivia et Patagonia").

Type.— M. C. Z. 77, 323, Lafr. coll. 4,607.

In addition to the cotype (?) of M. mentalis and the type of M. albi-

BANGS AND PENAKD: SOME CRITICAL NOTES ON BIRDS. 27

mentum, there is another specimen from Chile in the Lafresnaye Col-
lection, M. C. Z. 77,317, Lafr. coll. 4,600. These three specimens
are very much smaller than any of a large series from the Falkland
Islands in M. C. Z., from which we consider them subspecifically dis-
tinct. We therefore revive the name mentalis for the small continen-
tal form.

Sclater (Cat. birds Brit, mus., 1888, 14, p. 55) doubtfully places
M. albimentum Lafr. in the synonymy of M. cincrca Philippi and
Landbeck. The type is a specimen without the brown spot on the
chin, and we identify it with certainty as Muscisaxicola macloviana
mentalis d'Orb. and Lafr. of which it is presumably a female, as Lafres-
naye himself at first thought.

In designating our specimen of M. mentalis d'Orb. and Lafr. as a
cotype, we offer the following explanation : —

The types of the species originally described in joint authorship by
d'Orbigny and Lafresnaye are generally supposed to be in the Paris
Museum, and undovibtedly the specimens there have a just claim
to that distinction. We know, however, that many of these species
were represented by several individuals, and in such cases Lafresnaye
obtained from d'Orbigny a specimen, or several specimens, for him-
self. It is even possible that the descriptions were made from Lafres-
naye's specimens, since it is known that he did most of the work in
connection with the preparation of the manuscripts of papers published
in joint authorship "with d'Orbigny.

Under the circumstances, and in absence of positive proof to the
contrary, we think the specimens in the Lafresnaye collections should
be considered cotypes whenever there is sufficient information to
indicate that they were of d'Orbigny's collecting. Evidence of this
nature is to be found on the old labels in the form of a statement of
the locality, exactly as originally published, often accompanied by
d'Orbigny's name. In our proposed account of the Lafresnaye Col-
lection we shall enter such specimens as cotypes followed by a query
to indicate the doubtful status of the specimen, in the manner indi-
cated above. In all instances we shall give an exact copy of the
labels, so that ornithologists may decide for themselves upon the
merits of each individual case.

Muscisaxicola albilora albilora Lafresnaye.

Mnscisaxicola albilora Lafr., Rev. et mag. zool., 1855, p. i>0 (type-
locality not stated — we select Vicinity of Santiago, Chile).

28 bulletin: museum of comparative zoology.

Tyve.— M. C. Z. 77,322, Lafr. coll. 4,604.

Miiscisaxicola rubricapilla Philippi and Landbeck, Archiv. natur-
gesch., 1865, 1, p. 90 ("Cordillera der Hacienda de la Puerta, Provinz
Colchagua," Chile).

The specimen which proves to be the type of Lafresnaye's Musci-
saxicola albilora was listed by Verreaux in his Catalogue of the Lafres-
naye Collection as Muscisaxicola rufivertex, and by Sclater (Cat.
birds Brit, mus., 1888, 14, p. 58) doubtfully as a synonym of Musci-
saxicola jtminensis Tackzanowski. A comparison with M. juninensis
and M. rubricapilla shows bej'ond doubt its identity with the latter,
which it antedates by ten years. The two subspecies should now
stand as : —

Muscisaxicola albilora albilora Lafr.

Muscisaxicola albilora jtminensis Tackz.

Tachuris Lafresnaye.

Tachuris Lafresnaye, Echo du monde savant, 12 June, 1836, p. 107.

Type. — Regulus omnicolor Vieillot, Gal., 1834, 1, p. 271, pi. 166 =
Sylvia rubrigastra Vieillot, 1817.

Tachuris d'Orbigny and Lafresnaye, Mag. zool. Syn. avium, 1837,
p. 55.

Cyanotis Swainson, Class, birds, 1837, 2, p. 243.

Type. — Regulus ovinicolor Vieillot.

The generic name Tachuris Lafresna3e (1836), ha-\ang priority over
Cyanotis Swainson (1837), must replace it. The two forms contained
in this genus are : —

Tachuris rubrigastra rubrigastra (Vieillot).

Tachuris rubrigastra alticola (Berlepsch and Stolzmann).

Myiochanes ardosiacus cineraceus (Lafresnaye).

Tyrannula cineraceaLsdr., Rev. zool., 1848, p. 7 (Caracas, Venezuela).

Type.— M. C. Z. 83,338. Lafr. coll. 8,400.

Myiochanes ardosiacus polioptilus Todd, Ann. Carnegie mus., 1912,
8, p. 208 (Lagunita de Aroa, Estado Lara, Venezuela).

In some manner Lafresnaye's name Tyrannula cincracea, long ago
became applied to the South American black phoebe, and that bird

BANGS AND PENARD: SOME CRITICAL NOTES ON BIRDS. 29

ever since has figured as Sayornis cineracea (Lafr.) or Sayornis nigri-
cans cineracea (Lafr.).

Lafresnaye's description certainly does not fit the South American
black phoebe, but agrees exactly with his type-specimen now before
us, which is not a Sayornis but a Myiochanes, the wide bill, pale
lower mandible, and gray colors all being mentioned by Lafresnaye.

We give the following brief description of the type, possibly a little
faded by long exposure to light: —

First primary shorter than fifth and longer than sixth (from outside) .
Wing, 84; tail, 71; tarsus, 16; exposed culmen, 15. Crown well
crested, dusky; upper parts gray, about between mouse-gray and
deep mouse-gray of Ridgway, lores whitish; wings dusky, no conspicu-
ous wing-bands; tail dusky, outer web of outer rectrix whitish;
below mouse-gray, throat whitish and middle of abdomen white.
Most nearly like Myiochanes ardosiacus ardosiacus (Lafr.) but slightly
smaller, paler, and grayer in general coloration, with the throat whit-
ish, middle of belly white and outer web of outer rectrix whitish.

Saj/ornis cineracea of authors, not of Lafresnaj'e, must become

Sayornis nigricans latirostris (Cabanis and Heine).

Heleodytes pallescens (Lafresnaye).

Campylorhynchiis pallescens Lafresnaye, Rev. zool., 1846, p. 93
("Mexique" — error, type-locality unknown, we suggest S. W.
Ecuador).

Type.— M. C. Z. 76,137, Lafr. coll. 2,613.

Campylorhynchus pollidus Lafresnaye, Rev. zool., 1846, p. 94.

Type.— M. C. Z. 76,137, Lafr. coll. 2,613.

Campyhrhynclms paUiceps Lafresnaye Ms. Ridgway, Proc. Bos-
ton soc. nat. hist., 1887, 23, p. 384.

Type.— M. C. Z. 76,154, Lafr.. coll. 2,614.

Campylorhynchus halteatus Baird, Rev. Amer. birds, 1864, p. 97,
98, 103.

(Type from Babahoyo, Ecuador, ex Sclater, Cat. Amer. birds, 1862,
p. 16, species 102).

We do not hesitate to pronounce Campylorhynchus pcdlescens
identical with Campylorhynchus bcdteatus Baird of western Ecuador
and northwestern Peru. Lafresna\'e's t^'pe has two labels which read
respectively " Thryoi. pcdlidus ou pallescens .... (Florent Mexique ?) "
and "Camp, pallescens nob. rev. 1846, 93 (Mexique)." The doubt in

30 bulletin: museum of comparative zoology.

regard to the source of the specimen, indicated on the first of these
labels, is reflected in the guarded statement made by Lafresnaye in
the original description, "II nous a ete vendu comme du Mexique."

A careful examination of the type-specimen reveals that the narrow
bars and transverse markings on the breast, referred to by Ridgway
(loc. cit., p. 385), are on a few leathers which do not belong to the bird,
but which had been glued on by the taxidermist to cover some bare
spots. The feathers on these patches are wider and of a wholly difPer-
ent shape from those belonging to the bird. The bird's own breast-
feathers are spotted as in H. balteatus, and not barred or lined. We
cannot detect any difference in the width of the white and dusky
bands across the remiges from those in the specimens of H. balteatus
from Peru which we have before us, and with which the type of H.
pallescens agrees in size and very closely in all respects except that the
darker markings are paler and more grayish brown instead of black-
ish, due to fading from long exposure to the light.

Hylocichla minima minima (Lafresnaye).

Turdns viinimus Lafresnaye, Rev. zool., 1848, p. 5 (" Habitat ad
Bogotam, in Nova-Grenada").

Type.— M. C. Z. 76,498, Lafr. coll. 3,54L

Hylocichla aliciae hickncUi Ridgway, Proc. U. S. N. M., 6 April,
1882, 4, p. 377 (Shde Mt., Ulster Co., New York).

Apparently no ornithologist of the present generation had examined
the type of Turdns minimus Lafresnaye, until we recently did so. By
common consent the name has appeared in all modern works among
the synonyms of Hylocichla ustulata swainsoni (Cabanis). We were
therefore surprised upon comparing the type to find that not only is
it an Alice's Thrush and not a Swainson's Thrush, but that it is an
extreme example of the southern form of Alice's Thrush, always known
as Hylocichla aliciae bicknelli Ridgway. If the specimen really came
from Bogota as Lafresnaye thought it did, it is also the southernmost
record for the subspecies, which otherwise has not been found winter-
ing in South America. In order to be certain that our identification
might not be questioned, we have submitted the type to the following
American ornithologists, Messrs. Batchelder, Brewster, Faxon,
Oberholser, and Richmond, who all agree with us.

The two subspecies are: —

Hylocichla minima minima (Lafr.).

Hylocichla minima aliciae (Baird).

BANGS AND PENARD: SOME CRITICAL NOTES ON BIRDS. ' 31

TuRDUS NUDiGENis Lafrcsnayc.

Turdus nudujenis Lafresnaye, Rev. zool., 1848 (January), p. 4
(Caracas, Venezuela).

Type.— M.C. Z. 76,501, Lafr. coll. 3,551.

Turdus gymnophthahms Cabanis, Schomburgk's Reis. Brit. Guiana,
1848 (= 1849?) 3, p. 665 (British Guiana).

Turdus niuligcnis Lafresnaye certainly has priority over Turdus
gymnophthalmus Cabanis, even though the numbers of the Revue
zoologique may not have been issued in the months of which they
bear the dates. Hartlaub (Archiv. naturgesch., 1850, 2, p. 51) includes
Schomburgk's Reisen in his "Bericht iiber die vogel wahrend des
Jahres 1849." We find also that the volmue in ciuestion contains a
third list of subscribers which we have every reason to believe was
made out after the second list appearing in the second volume, which
is dated "bis ende Februar 1848."

Dr. Chas. W. Riclmiond writes (in lift.) that the third vohmie of
Schomburgk was probably published early in 1849.

Turdus rufopalliatus Lafresnaye.

Turdus rofopalliatus Lafresnaye, Rev. zool., 1840, p. 259 ("Monte-
rey en Californie," — error, we substitute Acapulco, Southwestern
Mexico).

Type — M. C. Z. 76,520, Lafr. coll. 3,568.

Merula flavirostris (not Turdus flavirosfris Horsfield, 1821) Swain-
son, Philos. mag., 1827, new ser., 1, p. 369 (Temascaltepec, Mexico).

This bird was collected by Leclancher on the voyage of the Venus.
The expedition stopped at Monterey, California, but the specimen
was probably not taken there. The name Turdus rufopalliatus, to
replace Merida flaiyirostris Swainson, must be used by all ornithologists
who, like ourselves, unite Planesticus and Turdus.

CossYPHA NivEiCAPiLLA niveicapilla (Lafresnaye).

Turdus niveicapillus Lafresnaye, Mem. Soc. acad. Falaise. Essai
nouv. man., 1838, p. 16 (Senegal).

Type — M. C. Z. 76,465, Lafr. coll. 3,938.

Cossypha verticalis Hartlaub, Verz. Hamb., 1850, p. 23.

32 bulletin: museum of comparative zoology.

The specimen which proves to be the type of Turdus niveicapiUus
Lafr. was entered by Verreaux in the Catalogue of the Lafresnaye
Collection as Bessonornis sivamsoni Bp. A careful comparison of
specimens proves that Lafresnaye's bird is what has been currently
called Cossypha verticalis Hartlaub. Lafresnaye's name must, of
course, be used for the species, antedating Hartlaub's by twelve years.

Races of Saltator Striatipictus Lafresnaye.

We have before us a series of 104 skins of Saltator striatipictus, a
careful study of which forces us to recognize six races, two of which
are here described as new. It is possible to find examples of one race
which agree very nearly with some of another, but the different races,
in series, stand out very definitely.

Young birds are more heavily marked and darker below than adults,
and old birds in breeding plmiiage are somewhat grayer above than
they are in winter or autumn.

L Saltator striatipictus furax, subsp. no v.

Type.— M. C. Z. 118,65L Adult d". Western Costa Rica:
near Boruca. 27 May, 1906. C. F. Underwood.

Characters. — Similar to Saliator striatipictus striatipictus Lafresnaye,
diftering in being slightly smaller, and much darker below; the stripes
on the under parts very wide and heavy and oUve-green; the dark
stripes cover the whole under parts often including the belly, and tend
to coalesce at the sides of the breast and sides of the neck. This form
is most nearly related to S. striatipictus isthmicus Sclater; it is of about
the same size, but much darker and greener below and more hea^^ly
striped. The very darkest specimens of isthmicus (possibly immature
birds) resemble closely the very palest examples of the new form.

Measurements.^ Type, adult cf ; wing, 90; tail, 82; tarsus, 22.5;
exposed culmen, 18.

Specimens examined. — Twenty-one from western Costa Rica;
Boruca, Lagato, and El General.

2. Saltator striatipictus isthmicus Sclater.

Saltator isthmicus Sclater, Proc. Zool. soc. London, 1861, p. 130
(Panama).

Type-locality. — Panama.

BANGS AND PENARD: SOME CRITICAL NOTES ON BIRDS. 33

Characters. — Similar to S. s. striatipictus but slightly smaller, the
under parts more greenish or yellowish, less purely white and the stripes
rather heavier and more olive greenish, less gra^ash.

Specimens exammed. — Twenty-one from Panama; near Panama
City and Loma del Leon.

3. Saltator striatipictus speratus, subsp. nov.

Type.— M. C. Z. 40,501. Adult cf • Sabago Island, Pearl Islands,
Bay of Panama. 6 April, 1904. W. W. Brown, Jr.

Characters. — This form is about the size of S. s. isthmicus and
slightly smaller than S. s. striatipictus; in color and markings it is
intermediate, i.e., it is slightly more yello^dsh or greenish below than
striatipictus and less heavily striped on the under parts than isthmicus.

If this were not an island form we would be inclined not to give it
a name, but to call it a connecting link between the two races.

Measurements. — Type, adult cf ; wing, 93; tail, 88; tarsus, 23;
exposed culmen, 18.

Specimens examined. — Forty from Pearl Islands; Sabago Island,
and San Miguel Island.

4. Saltator striatipictus striatipictus Lafresnaye.

Saltator striatipictus Lafr., Rev. zool., 1847, p. 73.

Type-locality. — Caly, Colombia.

Characters. — Size slightly larger than in the preceding forms;
under parts nearly white, very little tinged wath greenish or yellowish;
stripes on under parts fewer, narrower, and more grayish, less greenish
in color.

Specimens examined. — Fourteen, "from Colombia; Caly (one co-
type), "New Grenada," Jimenez, San Luis Bitaco Valley, La Maria
Dagua Valley, Santa Marta, and Trinidad. (The Trinidad bird
may represent still another form.)

5. Saltator striatipictus peruvianus Coiy.

Saltator peruvianus Cory, Publ. 190 Field mus. nat. hist, ornith.
ser., 1916, 1, p. 34.5.

Type-locality. — Hda. Limon, 10 miles west of Balsas, northern Peru.
Characters. — Much larger than S. s. striatipictus (wing in d^ 100-

34 bulletin: museum of comparative zoology.

104) ; under parts hea^dly striped with olive-green, much as in isthmicus,
but with the ground color whiter, much less greenish or yellowish.
Specimens examined. — Six from northwest Peru; Huancabamba.

6. Saltator striatipictus immaculatus Berlepsch and Stolzmann.

Saltator immaculatus Berlepsch and Stolzmann, Proc. Zool. soc.
London, 1892, p. 375.

Type-locality. — Peru; Lima.

Characters. — About the size of peruvianus, but with the stripes on
under parts very faint and confined to the sides, the whole median
under parts nearly immaculate whitish.

Specimens examined.— Two from " Coast of Peru." These were
collected on the voyage of the Venus, and are cotypes of Lafresnaye's
manuscript name Saltator albiventris.

I
Chlorophonia pyrrhophrys (Sclater).

Euphonia pyrrhophrys Sclater, Contr. ornith., 1851, p. 89 (Colom-
bia).

Tanagra {Euphonia) pretrei (not Tanagra pretrei Lesson, 1839
= Spiridalis pretrei pretrei (Lesson)) Lafresnaye, Rev. zool.,
1843, p.. 97 (Colombia).

Tijpe.— M. C. Z. 76,905, Lafr. coll. 2,816.

Lafresnaye's name for the Blue-capped green tanager being clearly
preoccupied by Lesson must give way to Sclater's later Euphonia
pyrrhophrys.

Tanagra aureata aureata Vieillot.

Tanagra aureata Vieillot, Enc. meth., 1823, 2, p. 782 (Paraguay).

Pipra cyanocephala (not Tanagra cyanocephala P. L. S. Miiller,
1776, = Tangara cyanocephala cyanocephala (P. L. S. Miill.)) Vieillot,
Nouv. diet. hist, nat., 1818, 19, p. 165 (Trinidad).

Tanagra nigricollis (not of Gmelin 1789) Vieillot, Nouv. diet. hist,
nat., 1819, 32, p. 412 (Brazil).

Tanagra chrysogaster Cuvier, Reger. avium, 1829, 1, p. 366, ex.
Undo bleue dore d'Azara (Paraguay).

BANGS AND PENARD: SOME CRITICAL NOTES ON BIRDS. 35

Tanagra aurcata Vieillot becomes the name by which this species
must be known. We recognize three geographical races, as follows : —

1. Tanagra aureata aureata Vieillot.

A large southern race, with darkest under parts — more chestnut
or orange and palest blue crown. Range extending north to Bahia.

2. Tanagra aureata intermedia (Chubb).

Euphonia nigricoUis iniermedia Chubb, Ibis, 1910, ser. 9, 4, p. 624.
A smaller race, slightly paler below, with a slightly more purplish
blue crown. Colombia, Venezuela, and Guiana.

3. Tanagra aureata pelzelni (Sclater).

Euphonia nigricoUis pchcini v. Berlepsch, Ms. Sclater, Cat. birds
Brit, mus., 1886, 11, p. 61.

A very distinct race, with yellow under parts (lacking the brownish
or orange tinge present in the other two). Western Ecuador.

Tanagra lauta lauta, nom. nov.

Euphonia hirundinacea (not Tanagra hirundinacea Lesson, Traite
d'ornith., 1831, p. 460 = Cypmagra hirundinacea (Lesson)), Bonaparte
Proc. Zool. soc. London, 1837, p. 117 (Guatemala).

Since there is no name in synonymy available for Bonaparte's
Euphonia, which ranges from Mexico to Nicaragua, we propose the
above.

Tanagra lauta proba, nom. nov.

Phonasca Gnatho (not Tanagra gnatho Lichtenstein, 1830 = Sal-
tator atriceps atriceps Lesson) Cabanis, Journ. orn., 1860, p. 335 (Costa
Rica).

The name by which the Costa Rican form has been known also
proves to be untenable, and finding no other applied to it, we propose
the above.

36 bulletin: museum of comparative zoology.

Tangara heinei Cabanis.

Procnias heinei Cabanis, Mus. Hein., 1850, 1, p. 31 (Colombia).

Tanagra (Aglaia) atricapilla (not Tanacjra atricapilla Gmelin, 1789)
Lafresnaye, Rev. zool., 1843, p. 290 (Colombia).

Type.— M. C. Z. 76,923, Lafr. coll. 2,931.

Since Lafresnaye's name for the Black-capped tanager is preoccupied
by Tanagra atricapilla Gmel., we take for the species the only other
name available in synonymy.

Iridosornis rufivertex rufivertex (Lafresnaye).

Arremon rufivertex Lafi'esnaye, Rev. zool., 1842, p. 335 excl. refer-
ence to Florent-Prevost (Bolivia).

Cotype.— M. C. Z. 76, 981, Lafr. coll. 2,951.

Cotype.— M. C. Z. 76,982, Lafr. coll. 2,950.

Tanagra duhusia Bonaparte, Consp. avium, 1850, 1, p. 239 (Colom-
bia).

In his original description of this tanager Lafresnaye referred to
"Florent-Prevost. zool. du voy. de la Venus," saying also "elle est
figure dans la voyage de la Venus." This was a mistake. The bird
Lafresnaye had in mind being Tanagra ruficcrrix Prevost and Des Murs
(now Tangara ruficervix (Prevost and Des Murs)). Later Lafresnaye
recognized his mistake in confusing his bird and Prevost's, and con-
sistently referred to the bird described by himself as "Nob.," crossing
out the reference to Prevost on the original label of his specimens. The
cotypes of Arremon rufivertex Lafr. are thus Lafresnaye's owaa speci-
mens having nothing to do with Tanagra ruficervix Prevost and Des
Murs, and Lafresnaye's name must supplant Bonaparte's duhusia
in current use for the species. With Lafresnaye's cotypes, original
labels, and description before us, we believe this change of names
correct. If, however, our disposition of names be considered errone-
ous, then the name Iridosornis Lesson must go. Lesson (Echo du
Monde Savant, 1844, p. 80) in specifying the type of his genus states
that — " Le type de ce genre, bien distinct dans la tribu des tangaras,
a ete decrit par M. Florent Prevost sous le nom' d' Arremon rufivertex
{Zool. de la Venus et Revue zool., 1842, p. 335)," evidently meaning
Lafresnaye's bird and not Tanagra ruficervix Prevost and Des Murs.
If, however, the type of the name Arremon rufivertex is not Lafresnaye's

BANGS AND PENARD: SOME CRITICAL NOTES ON BIRDS. 37

specimen, but Prevost's Tanagra ruHcervix, the latter must also be the
type of Lesson's genus Iridosornis! Poecilornis Hartlaub, 1844,
would be untenable for exactly the same reason, and we should have
to use Euthraupis Cabanis, 1850.

According to our views the three subspecies are : — ■

Iridosornis rufiveriex rufivertex (Lafresnaye).

Iridosornis rufivertex ignicapillus Chapman.^

Iridosornis rufivertex caeruleoveniris Chapman.'^

Tachyphonus surinamus brevipes Lafresnaye.

Tachyphonus brevipes Lafresnaye, Rev. zool., 1846, p. 206 (Colom-
bia).

Cotype — M. C. Z. 76,728, Lafr. coll. 3,100.

Cotype.— M. C. Z. 76,729, Lafr. coll. 3,101.

Tachyphonus napensis Lawrence, Ann. Lye. nat. hist. N. Y., 1864,
8, p. 42 (Rio Napo, East Ecuador).

Berlepsch, Rev. tanag. (Int. ornith. kongress, 1910, p. 1, 148),
listed Tachyphonus brevipes Lafresnaye among undetermined species,
suggesting that it might be the female of Tachyphonus surinamus
(Linne).

The two cotypes of T. brevipes Lafr. are adult females. They are
alike, each showing to a marked degree the ochraceous buff throat
and breast characteristic of the female of the form we have been
calling T. surinamus napensis La"WTence. The female of T. suri-
namus surinamus (Linne) has the throat and breast cream buff.
Lafresnaye's specimens came from Colombia and the name Tachy-
phonus surinamus brevipes Lafr. must therefore replace Tachyphonus
surinamus napensis LawT.

Chlorospingus ophthalmicus (Du Bus).

Arremon ophthalmicus Du Bus, Bull. Acad. Bruxelles, 1847, 14,
2, p. 106 (Mexico). •

Tachyjihonus albitempora (not of authors) Lafresnaye, , Rev. zool.,
1848, p. 12 ("Habit in Colombia," — error, we suggest Mexico).

Cotype.— M. C. Z. 77,050, Lafr. coll. 3,122.

Cotype — M. C. Z. 77,051, Lafr. coll. 3,123.

1 Bull. Amer. mus. nat. hist., 1915. 34, p. 656.

2 Loc. cit., p. 657.

38 bulletin: museum of comparative zoology.

Lafresnaye's two cotypes do not belong to the species with which
his name albitempora has always been associated, but are perfectly
characteristic examples of the Mexican ophthalmicus of Du Bus, and
probably came from southeastern Mexico. In many instances
Lafresnaye did not know whether his specimens were from Colombia
or Mexico, and we find numerous labels written by him which say,
"Colombie ou Mexique." At some date later than his description of
Tachyphonus albitempora, Lafresnaye himself thought his bird identi-
cal with Arremon ophthalmicus and wrote a second label for his speci-
mens to that effect.

Chapman (Bull. Amer. mus. nat. hist., 1917, 36, p. 618) while work-
ing on the Colombian forms of Chlorospingus, appears to have been
the first ornithologist of the present generation to detect the absolute
discrepancy between Lafresnaye's description and the Colombian
bird to which the name had universally been applied. He therefore
named the Colombian form Chlorospingus albitempora nigriceps.

We suppose the type of Chlorospingus flaviventris Sclater is in the
Museum of Cambridge LTniversity; it should be examined and com-
pared because if, as supposed by Salvin, it represents what was known
as C. albitempora Lafr., it bears the earliest date of any of the sub-
species. Trinidad, whence it was supposed to come, is undoubtedly
an error, and the subspecies to which it belongs must be proved before
a new arrangement of the forms of this species can be made.

Cnemoscopus, gen. nov.

Type. — Arremon rubrirostris Lafresnaye.

Characters. — Similar to Hemispingus in form and in shape of bill;
legs much shorter - — wing four and one quarter times the length of
the tarsus (three and one half times in Hemispingus); coloration
decidedly different from any of the species in the genus Hemispingus,
the red bill, gray head, and yellowish green body being very distinc-
tive. Except for the more slender, red bill, the general appearance
suggests the genus Eucometis.

OsTiNOPS decumanus insularis Dalmas.

In 1900 (Mem. Soc. zool. France, 13, p. 137) Count Dalmas named
the Great yellow-tail of Tobago, basing his separation upon the smaller
size and paler castaneous rmnp of the island form. In 1906, Hell-

BANGS AND PENARD: SOME CRITICAL NOTES ON BIRDS. 39

mayr (Nov. zool., 1906, 13, p. 19) criticised Dalmas's form, and viewed
the separation as a mistake, on the ground that the characters given
by Dalmas were precisely those distinguishing the female from the
male of the species, and suggested that the specimens seen by Dalmas
were incorrectly marked as to sex.

In 1917 Todd (Proc. Biol. soc. Washington, 30, p. 3), on the other
hand, named the Colombian form Ostinops decumamis melmiterus
supposing the Colombian bird to be blacker than the Guianan.
Chapman (Bull. Amer. mus. nat. hist., 1917, 36, p. 24) has entirely
disproved this, showing that there is no difference in color between
Colombian and Guianan specimens. He, however, noticed the more
chestnut tone of birds from Trinidad and the Paria Peninsula.

We have lately examined and compared a large series from the con-
tinent and from Trinidad and Tobago, and while we, like Chapman,
cannot find any differences in specimens from Colombia and Guiana,
we believe that the paler coloration and castaneous upper parts,
especially the rump, of birds from Tobago, Trinidad, and the Paria
Peninsula (the latter on Chapman's authority), are constant char-
acters, and we therefore re\'ive the name Ostinops decumanus insularis
Dalmas.

CiSSILOPHA SANBLASIANA SANBLASIANA (Laf resuayc) .

Pica san-blasiana Lafr., Mag. zool., 1842, pi. 28 ("Elle vit en
troupes selon M. Leclancher a Acapulco et a San-Blas sur la cote
ouest du Mexique" — we select Acapulco).

Type.— M. C. Z. 76,202, Lafr. coll. 5,543. Acapulco.

Cissolopha pulchra Nelson, Auk, 1897, 14, p. 56 (Acapulco, Guerrero,
S. W. Mexico).

Although Lafresnaye, in naming this species, cited " Geai de San-
Blas, Neboux, Rev. zool., 1840, p. 290, et 323," he described from his
own specimen and even called attention to the fact that his bird was
slightly different from Neboux's. Lafresnaye's bird, the type of the
species, came from Acapulco as stated by Lafresnaye (Rev. zool.
1840, p. 323) in quoting Leclancher from whom he obtained it: " Cette
Pie noire et bleue vient d' Acapulco."

Dr. E. W. Nelson, has redescribed this southern race, as Cissolopha
pulchra, assuming the type-locality of Pica sanblasiana to be San Bias.
Dr. Nelson has kindly lent us the type of C. pidchra and a long series
of topotypes. These we have compared with more than fifty skins
from Cohma, Tepic, in M. C. Z.

40 bulletin: museum of comparative zoology.

Lafresnaye's type certainly represents the southern form, with the
blue of a much more purplish shade than in any example from Colima.
It is not quite so dark as Nelson's type, which is an extreme example,
but compared with a series of topotypes it is an average specimen.
The specimen was mounted and on exhibition, in direct light, in the
Boston Society of Natural History for more than fifty years ; it shows,
however, but slight injury; the blue is perhaps a little dulled and the
black has become somewhat brownish.

It is now obvious that C. s. pulchra Nelson is a synonym of C. s.
sanhlasiana (Laf resnaye) . The northern form being without a name,
we take pleasure in naming it in honor of Dr. Nelson.

CissiLOPHA sanblasiana nelsoni, subsp. nov.

Tyye — M. C. Z. 65,111. Adult d". Mexico:' Colima. 21 March,
1913. Gustav. Gliickert.

Characters. — Similar to C. s. sanhlasiana (Lafresnaye) of Acapulco,
but smaller; upper parts bright cerulean blue instead of rich ultra-
marine or cyanine blue; under tail coverts and thighs dull ultra-
marine blue instead of cyanine blue.

Measurements. — Type, adult cf; wing, 143.5; tail, 155; tarsus, 39;.
exposed culmen, 31.

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.
Vol. LXIII. No. 3.

THE ANTS OF BORNEO.

By William Morton Wheeler.

CAMBRIDGE, MASS., U. S. A.

PRINTED FOR THE MUSEUM.

July, 1919.

No. 3. — The Ants of Borneo.

By William Morton Wheeler.

contributions from the entomological laboratory of the
bussey institution of harvard university, no. 145.

During the past decade several collections of Bornean ants have
been sent me for study and identification. Mr. John Hewitt sent an
interesting lot of specimens accumulated diu-ing his residence in Kuch-
ing and Prof. Harrison W. Smith, of the Massachusetts Institute of
Technology, made a collection in the same locality for the Museum of
Comparative Zoology. He also contributed a number of specimens
collected in British North Borneo by Mr. E. B. Kershaw, a clever
young naturalist who lost his life in that country in a forest fire. Prof.
Roland Thaxter of Harvard University gave me a number of small
species from Sarawak, and Mr. Horace Donisthorpe kindly sent
several that had been taken by Mr. G. E. Bryant on Mt. Matang,
near Kuching. Recently a few additional specimens were received
from Mr. William Beebe, of the New York Zoological Park.

While working up this material I found it necessary to prepare a
complete list of the known Bornean Formicidae with their more
important synonjTtiy and distribution. During recent years less
attention has been bestowed on the ants of Borneo than on those of
Java, Sumatra, the Malay Peninsula, Burmah, and India. The
Bornean fauna has, however, considerable historical interest to the
taxonomist, because it has been studied by all the leading myrme-
cologists, Smith, Mayr, Ernest Andre, Emery, and Forel, and because
the researches of several of these investigators were based on material
secured by such well-known collectors and explorers as Alfred Russel
Wallace, Doria, Beccari, Bedot, Pictet, and Chaper. Thus Borneo
has come to be the type-locality for many interesting species later
found to have a wade distribution in Indonesia. The material sent
me by Hewitt and Harrison W. Smith is valuable because it was
taken in the very localities in which Wallace collected. Professor
Smith has given me the following notes in regard to some of these:

" Seravihu is the little mountain a few miles up country from
Kuching, on the Sarawak River, where the first Rajah had his bunga-

44 bulletin: museum of comparative zoology.

low and where Wallace made the celebrated collection of moths to
which he refers in "The Malayan Archipelago."

" Rambungan River is a small stream entering the sea about ten miles
south of the Sarawak River. The specimens were taken about twelve
miles from the coast.

" Sadcnig is the town on the Sadong River where the first coal mine
was started.

" Matang Mountain is the beautiful mountain which one sees from
Kuching."

The total number of species of which I have been able to make a
record from Borneo is 256. I may have overlooked a few, owing to the
widely scattered publication of the original descriptions and citations
of localities. On the whole, the fauna has many forms in common with
Sumatra, Java, and the Malay Peninsula, and additional exploration
will no doubt greatly increase the number of such species. Quite a
number of forms, however, seem to be peculiarly Bornean. The
total number of genera is 59, distributed among the five subfamilies
as follows: Ponerinae 18, Dorylinae 2, Myrmicinae 23, Dolichoderinae
4, Camponotinae 12. The following pages add some 58 species
(indicated by an asterisk) to the known fauna, including 23 new to
science.

The series of Bornean ant genera comprises several of ancient aspect,
e.g., Cerapachys, Phyracaces, Metapone, Acanthomyrmex, Calyp-
tomyrmex, Rhopalothrix, Cataulacus, Myrmoteras, Aphomomyrmex,
Oecophylla, Gesomyrmex, Dimorphomyrmex, and Echinopla. Some
of these seem to be confined to the mountains of Borneo and to be
represented also in the mountains of Burmah and the Philippines.
One species, Gesomyrmex chaperi, is unusually interesting, as it is
peculiar to Borneo. The genus was first described by Mayr in 1868
from the Baltic Amber, and many years elapsed before the living
Bornean species was discovered by Ernest Andre (1892). In the
same paper Andre described a species of Dimorphomyrmex from
Borneo and three years later Emery recorded a fossil species from the
Baltic Amber. Recently I described a second living species from the
mountains of Luzon. It is probable, therefore, that careful explora-
tion of the mountains of Borneo and the neighboring islands will
bring to light other interesting relicts of the once very widely dis-
tributed Eocene ant-fauna.

Some of the Bornean ant-genera are very rich in species, e.g.,
Crematogaster, which is remarkable also in comprising more numer-
ous forms with 10-jointed antennae (subgen.' Decacrema) and swollen

wheeler: the ants of Borneo. 45

epinotum (siibgen. Physocrema) than occur in other regions. Lepto-
genys is represented by several large and handsome species of the
subgenus Lobopelta, and Cataulacus, Dolichoderus (subgen. Hypo-
clinea), Polyrhachis, and Echinopla are also rich in species. All but
three of the subgenera of Polyrhachis (Hagiomyrma, Hedomj^ma,
and Myrmatopa) are known to occur in the island. Of Camponotus
the subgenera Myrmotarsus and Colobopsis are represented by
numerous species, while most of the other subgenera are rather poorly
represented, though often by peculiar forms (Myrmoturba, Myrmo-
sphincta). Compared with the Philippines and the adjacent main-
land, and especially with Papua and Australia, Borneo seems to possess
few species of Pheidole and Monomorium, and many primitive poner-
ine genera have not been recorded from the island, e.g. Mystrium,
Stigmatomma, Trapeziopelta, Prodiscothyrea, Cryptopone, and
Centromyrmex. I believe, however, that some or all of these will
be foimd in Borneo. Only recently I received species of Mystrium,
Stigmatomma, Trapeziopelta, and Centromyrmex from the Phil-
ippines, where they were previously unknown. Of course, Borneo
has been invaded by the usual tropicopolitan tramp species, Mono-
morium pharaonis and floricola, Telraviorium guineense and similli-
mum, Pheidole viegacephala, Triglyphothrix striatidens, Plagiolepis
longipes, and Prenolepis longicornis, and obscura..

FORMICIDAE : PONERINAE.

1. Cerapachys antennatus Smith.

Cerapachys antennatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 74, y ; Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5, p. 445, 9, pi. 1,
fig. 8, 9; Emery, Gen. Ins. Ponerinae, 1911, p. 9, S 9.

Type-locality: Sarawak, Borneo (A. R. Wallace).
Worker. Length 5.5 mm.

Head distinctly longer than broad, narrower in front, than behind, with
straight sides, broadly concave occipital border, acute occipital angles and
convex dorsal surface, subtruncate behind. Eyes rather large, flattened,
their anterior orbits at the middle of the head. There is a small shallow
impression on the middle of the vertex. Occipital border marginate, the
margination surrounding the corners and continued some distance along the
ventral surface of the head. Gula with a pair of small, prominent teeth at

46 bulletin: museum of comparative zoology.

its anterior margin. Mandibles rather large, triangular, strongly bent at the
base, flattened, with straight lateral and broad, dentate apical borders. Clyp-
eus extremely short and transverse. Frontal carinae prominent, vertical,
approximated and rounded, confluent but not truncated behind in a depression
continuous with the antennal foveae. In front between the carinae there is a
small, acute, median tooth. Frontal groove absent. Cheeks with a strong
carina, terminating anteriorly in a sharp, rectangular tooth or projection.
Antennae short; scapes rapidly enlarging towards their apices, which reach
back to a line connecting the anterior orbits; funiculi thick, all the joints
except the last decidedly broader than long, joints 1-6 much broader than long,
joints 7-10 subequal, somewhat longer, terminal joint very large, glandiform,
as long as the four preceding joints together. Thorax narrower than the head,
about 2 1 times as long as broad, as broad through the epinotum as through
the pronotum, narrowed in the mesoepinotal region; with indistinct, slightly
impressed mesoepinotal suture. Pronotum subrectangular in front, its ante-
rior and inferior borders strongly marginate. In profile the dorsal outline
of the thorax is horizontal and very feebly convex. Epinotum from above a
little longer than broad, rather rounded on the sides, its declivity sloping,
slightly concave and strongly marginate above and on the sides. Petiole
narrower than the epinotum, distinctly longer than broad, as broad in front
as behind, with feebly rounded sides and dorsal surface, the former slightly
carinate below; its ventral surface anteriorly with a prominent, compressed,
triangular tooth. The anterior surface is strongly truncated and with a sharp
carina above. Postpetiole a little longer than the petiole but distinctly
broader, longer than broad and broader behind than in front, with evenly
convex dorsal, ventral, and lateral surfaces, its anterior border strongly margin-
ate, with sharply angular corners. Gaster elongate, first segment shaped
like the postpetiole but larger; pygiclium truncated and slightly concave
above, bordered with numerous prominent spinules. Sting large. Legs with
short tibiae, those of the middle and hind legs not longer than the rbetatarsi;
hind coxae without a lamelliform expansion at the tip on the inner side.

Shining; mandibles opaque, striatopunctate; cheeks very coarsely rugose.
Surface of the body with very sparse, coarse, piligerous punctures, longitudi-
nally confluent on the dorsolateral surfaces of the petiole.

Hairs moderately long, bristly, erect, pale yellowish, sparse on the body,
sparser on the scapes and legs. Pubescence absent, except on the funiculi
tibiae, and tarsi.

Black; mandibles, funiculi, tarsi, tips and bases of scapes, femora, and
tibiae, pygidium and sting deep red.

A single specimen from Kuching (John Hewitt).

I have redeseribed this insect which is the type of the genus, as the
worker has not been seen within recent years and because Smith's
description is antiquated and incomplete. Emery described and

wheeler: the ants of borneo. 47

figured a dealated female from Sumatra. Compared with his figures,
the worker has the petiole distinctly longer and more narrowed in
front, and the head is also narrower anteriorly.

2. Cerapachys dohertyi Emery.

Cerapachys dohertyi Emery, Rend. R. accad. sci. Bologna, 1901, p. 25, S ;
Emery, Gen. Ins. Ponerinae, 1911, p. 9, ^ .

Type-locality: Pulo Laut, Borneo (W. Doherty).

3. Cerapachys parvulus Emery.

Cerapachys dohertyi var. parvula Emery, Rend. R. accad. sci. Bologna, 1901,
p. 25, ^ ; Emery, Gen. Ins. Ponerinae, 1911, p. 9.

Type-locality: Pulo Laut, Borneo (W. Doherty).
As Emery surmised, this is, in all probability, a distinct species and
not a variety of dohertyi.

*4. Cerapachys bryanti, sp. nov.

Worker. Length 2 mm.

Head about \ longer than broad, a little broader behind than in front, with
very feebly convex sides, broadly excised and marginate posterior border and
sharp posterior angles. Eyes distinctly smaller than the greatest diameter of
the scapes, their posterior orbits at the middle of the head. Cheeks with a
prominently angled carina in front. Mandibles small, not flexed at the base,
with feebly rounded lateral and very indistinctly denticulate apical borders.
Frontal carinae prominent, erect, approxim'ated, rounded, subtruncate, but
not fusing behind. Frontal groove absent. Antennal scapes thick, about
half as long as the head, joints 1^10 of the funiculus very short and transverse,
terminal joint large, glandiform, as long as the six preceding joints together.
Thorax narrower than the head, elongate, subrectangular, flattened above
and on the sides, twice as long as broad, not broader behind than in front,
slightly narrowed in the middle, without promesonotal or mesocpinotal sutures.
Anterior border of pronotum very straight and transverse, its superior, and
inferior borders as well as the superior and lateral borders of the abrupt
epinotal declivity marginate. Petiole nearly square, very slightly broader
behind than in front, narrower than the epinotum, truncated and sharply
marginate in front, but not on the sides. Postpetiole shaped exactly like the
petiole, but larger. Gaster formed very largely of the first segment, which

48 bulletin: museum of comparative zoology.

has the same shape as the postpetiole but is somewhat larger. Terminal
gastric segments small. Pygidium minutely and rather bluntly spinulate
on the sides. Sting well-developed. Legs short and robust, hind coxae
without a lamelliform enlargement at the tip on the inner side.

Shining; head, thorax, petiole, and postpetiole evenly covered with sparse,
coarse umbilicate, piligerous punctures or foveolae, excepting the middorsal
portion of the thorax, which is smooth and very shining. Gaster sparsely
punctate, more finely than the more anterior regions, legs and scapes with
sparser, finer punctures. •

Hairs pale yellow, sparse, erect, bristly, of uneven length, less numerous on
the appendages than on the body. Pubescence absent, except on the funiculi.

Castaneous; mandibles, antennae, pygidium, sting, and legs, excluding the
coxae, red.

Described from a single specimen sent me by Mr. Horace Donis-
thorpe. It was taken by Mr. G. E. Bryant on Mt. Matang in West
Sarawak.

This species has the appearance of a Syscia on account of its small
size and the structvire of the thorax and abdomen, but the antennae
are 12-jointed as in Cerapachys sens. str. It is allied to C. dohertyi
Emery and parvula Emery, but both of these forms are decidedly
larger and have the petiole and postpetiole broader than long.

5. Phyracaces pubescens Emery.

Phyracaces puhescens Emery, Rend. R. accad. sci. Bologna, 1901, p. 26, 9 ;
Emery, Gen. Ins. Ponerinae, 1911, p. 11, 9.

Type-localiiy : Pulo Laut, Borneo (W. Doherty).

*6. Phyracaces hewitti, sp. nov.

Worker. Length 3.5 mm.

Head a little longer than broad, scarcely broader behind than in front, with
feebly rounded sides, broadly excavated posterior border and sharp posterior
corners, both strongly marginate. In profile the dorsal surface is moderately
convex, subtruncate behind, the gular surface feebly convex. Eyes rather
large, feebly convex, distinctly in front of the middle of the head. Mandibles
triangular, strongly bent at the base, with nearly straight external and in-
distinctly denticulate apical borders. Frontal carinae approximated, erect,
rounded, united but not truncated behind. Cheeks with a short, strong
carina, terminating in front in an acute, rectangular tooth. Antennal scapes

wheeler: the ants of borneo. 49

thickened towards their tips, which extend a httle beyond the posterior orbits;
fiinicuU rather long, joints 1-9 broader than long, joint 10 as long as broad,
terminal joint as long as the two preceding joints together, somewhat tapering
and not broader than the penultimate joint. Thorax shghtly narrower than
the head, distinctly broader through the epinotum than through the pronotimi,
less than twice as long as broad, without promesonotal and mesoepinotal
sutures. Pro- and mesonotum together rectangular, as long as broad; epino-
tum with rounded, rather swollen sides. In profile the whole thorax is feebly
and evenly rounded above. Epinotal declivity abrupt, very strongly carmate
above and on the sides, as is also the pronotum. Pleurae concave. Petiole
as broad as the epinotimi, rectangular, nearly I5 times as broad as long, as
broad in front as behind, feebly convex above, truncated anteriorly and
posteriorly, with its anterior and lateral borders marginate and its posterior
angles produced as a pair of triangular, rather acute teeth. Postpetiole
rectangular, a little broader than long, as broad in front as behind and as
broad as the petiole, feebly convex above, marginate in front, with sharp
anterior corners, submarginate on the sides. First gastric segment a httle
larger than the postpetiole, as long as broad, with more convex sides and
dorsum, ventrally, in front, with a conspicuous rounded tubercle. Pygidium
truncate, with finely spinulate border. Legs rather short, hind coxae with a
rounded expansion at the tip on the inner side.

Moderately shining; mandibles very sparsely and coarsely punctate; body
finely punctate, dorsal surfaces of head and thorax also with irregular scattered
foveolae; region between the eyes and frontal carinae smooth and shining.
Sides of head and thorax also more shining and less punctate.

Hairs and pubescence grayish, the hairs rather short, sparse, erect, both on
the body and appendages, longest and most abundant at the tip of the gaster,
the pubescence rather long and abundant, especially on the petiole, postpetiole,
gaster, legs, and antennae, but also well-developed on the thoracic dorsum and
head.

Black; mandibles, antennae, legs, pygidium, and sting dark red, the middle
portions of the scapes, femora, and tibiae somewhat darker.

Female. Length 4 mm.

Very similar to the worker. Thorax through the wing-insertions- as broad
as the head; mesonotum small, flat, a httle broader than long, shaped like an
isosceles triangle, with the apex directed anteriorly. Sculpture, pilosity, and
color as in the worker. Wings yellowish hyaline, with pale yellow veins and
conspicuous brown pterostigma.

Described from four workers and three females taken by Mr. John
Hewitt at Kuching. Type.— M. C. Z. 8,945.

This species seems to be closely related to Ph. pubescens Emery,
described from a dealated female, but hewitti is much smaller {pubes-

50 bulletin: museum of comparative zoology.

cens measures about 6 mm.) ; the epinotal declivity is very distinctly
separated from the base by a pronounced margination or carina, the
petiole, and postpetiole are much broader and the former has distinctly
dentate posterior angles.

*7a. Myopopone castanea Smith subsp. maculata Roger.

Myopopone ?naculata Roger, Berl. ent. zeitschr., 1861, 5, p. 50, § 9 .
Myopopone castanea Forel, Joui'n. Bombay nat. hist, soc, 1900, 13, p. 54, ^ 9 ;

Bingham, Fauna Brit. India. Hymenop., 1903, 2, p. 54.
Myopopone castanea suhsp. 7naculata Emery, Gen. Ins. Ponerinae, 1911, p. 26.

Type-locality: Ceylon ( ^ ) and Bintang Island (9 )•
A single female from Kuching (John Hewitt), though measuring
only 12.5 mm., agrees in ail other respects with females from the
Philippines. The tibiae are not spotted. The species has not been
recorded from Borneo, though well known from other parts of the
Malayan and Papuan regions.

*8. Platythyrea pusilla Emery.

Platythyrea pusilla Emery, Rev. Suisse zooL, 1893, 1, p. 188, ^ ; Emery,
Gen. Ins. Ponerinae, 1911, p. 29, S .

Type-locality: Amboina.

A single dealated female from Kuching (John Hewitt) agrees well
with Emery's description of the worker. It measures only 5 mm.

9. Platythyrea subtilis Emery.

Platythyrea subtilis Emery, Ann. Mus. civ. Geneva, 1900, ser. 2, 20, p. 666,
nota ^ ; Emery, Gen. Ins. Ponerinae, 1911, p. 29, § .

Type-locality: Pulo Laut, Borneo (Doherty).

10. Stictoponera borneensis Emery.

Ectatomma coxale Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 150, ^ {nee Roger).
Stictoponera borneensis Emery, ibid., 1900, ser. 2, 20, p. 662 nota, ^ ; Emery,
Gen. Ins. Ponerinae, 1911, p. 47, ^ . .

Type-locality: Sarawak, Borneo.

Two workers from Kuching (John Hewitt).

wheeler: the ants of Borneo. 51

Ua. Stictoponera costata Emery var. unicolor Forel.

Stictoponera costata var. unicolor Forel, Rev. Suisse zool., 1901, 9, p. 335, ^ ;
Emery, Gen. Ins. Ponerinae, 1911, p. 48, § .

Type-locality: Sarawak, Borneo (Haviland).

12. Stictoponera coxalis (Roger).

Ponera coxalis Roger, Berl. ent. zeitschr., 1860, 4, p. 308, ^ .
Ectatomma coxale Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5, p. 444.
Stictoponera coxalis Emery, ibid., 1900, ser. 2, 20, p. 662; Emery, Gen. Ins.

Ponerinae, 1911, p. 48.
Ectatomma (Stictoponera) coxale Bingham, Fauna Brit. India. Hymenop., 1903,

2, p. 84, ^ , fig. 44.

Type-locality: Ceylon (H. Nietner).
Sarawak (Doria and Beccari).

13. Stictoponera menadensis Mayr.

Ectatomma {Stictoponera) menadensis Mayr. Verb. Zool. bot. gesellsch. Wien,

1887, 37, p. 539 nota Q .
Stictoponera menadensis Emery, Ann. Mus. civ. Genova, 1900, ser. 2, 20,

p. 663; Emery, Gen. Ins. Ponerinae, 1911, p. 48, ^ .

Type-locality: Menado, Celebes (Radoszkowski).

Borneo.

A worker, which agrees very closely with Mayr's description, and
thi-ee males from Kuching (John Hewitt). I refrain from describing
the latter, as I am not certain that they belong with the worker.

14. Stictoponera rugosa (Smith),

Ponera rugosa Smith, Journ. Proc. Linn. soc. London. Zooh, 1857, 2, p. 66, S .
Stictoponera rugosa Emery, Gen. Ins. Ponerinae, 1911, p. 48, ^ .

Type-locality: Sarawak, Borneo (A. R. Wallace).

52 bulletin: museum of comparative zoology.

15. Rhopalopone diehli (Forel).

Edatomma (Mictoponera) diehli Forel, Ann. Soc. ent. Belgique, 1901, 45,

p. 372, ^ .
Rho-palopone diehli Emery, Gen. Ins. Ponerinae, 1911, p. 35, ^ .

Type-locality: Sarawak, Borneo (Haviland).

16. Odontoponera transversa (Smith).

Ponera transversa Smith, Jom-n. Proc. Linn. soc. London. Zool., 1857, 2,
p. 68, S ; Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 86, ^ .

Ponera denticulata Smith, ibid., p. 90, 9 , pi. 6, fig. 13, 14.

Odontoponera denticulata Mavi', Verh. Zool. bot. gesellsch. Wien, 1862, 12,
p. 717.

Odontoponera transversa Dalla Torre, Cat. Hymenop., 1893, 7, p. 30; Emery,
Gen. Ins. Ponerinae, 1911, p. 60.

Type-locality: Singapore (A. R. Wallace).

Sarawak (Doria and Beccari).

Several specimens from Kuching (John Hewitt), Rambungan River,
Sarawak (H. W. Smith) and British North Borneo (E. B. Kershaw).
These and a series of specimens taken at Surubaya, Java, by H. W.
Smith and by F. X. Williams in the Philippines vary considerably
in size. Rather pronomiced differences in color have also been noted
by other authors, but no attempt has been made to name varieties.

17. Diacamma holosericeum (Roger).

Ponera holoserica Roger, Berl. ent. zeitschr., 1860, 4, p. 302, ^ .

Diacamma holosericeum Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 149, ^ ;

Emery, ibid., 1887, ser. 2, 5, p. 435, ^ ; Emery, Gen. Ins. Ponerinae,

1911, p. 65.

Type-locality: Java.
Sarawak (Doria and Beccari).

18. Diacamma intricatum (Smith).

Ponera intricata Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 67, ^ .
Diacamma intricatum Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 149; Emery,
Gen. Ins. Ponerinae, 1911, p. 65.

wheeler: the ants of Borneo. 53

Type-locality: Sarawak, Borneo (A. R. Wallace).

Sarawak (J. Doria and O. Beecari); Kapouas Basin (Chaper);
Central Borneo (Munich Museum).

Several workers taken at Kuching by John Hewitt and H. W.
Smith and by the latter at Sadong, Serambu Mt., and Rambungan
River, Sarawak.

*18a. DiACAMMA INTRICATUM subsp. KERSHAWI, Subsp. nOV.

Worker. Length about 11 mm.

Smaller than the typical iniricaium, with smaller eyes, less pronounced
cljrpeal carina, the petiole narrower anteriorly, and the striae on the upper
surface of the petiole and first gastric segment almost obliterated. The
second gastric segment is faintly, but distinctly, longitudinally striated in the
middle above, and the whole gaster is more opaque and more densely punctate.
The pilosity on the body and legs is slightly more abundant than in the typical
form.

A single specimen taken by Mr. E. B. Kershaw in British North
Borneo. Type.— M. C. Z. 9,073.

19. DiACAMMA RUGOSUM (Le Guillou).

Ponera rxigosa Le Guillou, Ann. Soc. ent. France, 1840, 10, p. 318, S .

Ponera versicolor Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 65, ^ .

Diacamma rugosum Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 149, ^ ; Emery,

Gen. Ins. Ponerinae, 1911, p. 66.
Diacamvia geometricum subsp. versicolor Emery, Rev. Suisse zool., -1893, 1,

p. 189; Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5, p. 438.

Type-locality: Borneo. (Voyage of the "Astrolabe" and "Zelee").

Sarawak (Doria and Beecari; Bedot and Pictet; A. R. Wallace).

Eleven workers from Sadong, Sarawak (H. W. Smith) and two from
Kuching (John Hewitt) agree with Emery's redescription of this
species. They lack metallic reflections, however.

19a. DiACAMMA RUGOSUM Subsp. TORTUOLOSUM (Smith).

Ponera tortuolosa Smith, Journ. Proc. Linn. soc. London. Zool., 1863, 7, p. 18, t^
Diacamma tortuolosuni Forel, Mitth. Naturh. mus. Hamburg, 1901, 18, p. 46, U

54 bulletin: museum of comparative zoology.

Diacamma rugosum subsp. tortuolosmn Emery, Rend. R. accad. sci. Bologna,
1897, 1896-97, n. s., 1, p. 160; Emery, Gen. Ins. Ponerinae, 1911, p. 07.

Type-locality: Ceram (A. R. Wallace).
Tandjong, S. E. Borneo (Fritz Suck).

19b. Diacamma rugosum subsp. geometricum (Smith).

Ponera geometrica Smith, Journ. Proc. Linn. see. London. ZooL, 1857, 2,

p. 67, g .
Diacamma geometricum Mayr, Verb. Zool. bot. gesellsch. Wien, 1862, 12,

p. 718, S .
Diacamma javanum Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5, p. 439, ^ .
Diacamma rugosum subsp. geometricum Emery, Rend. R. accad. sci. Bologna,

1897, 1896-97, n. s., 1, p. 154, fig. 3, 8, 14, ^ ; Emery, Gen. Ins. Ponerinae,

1911, p. 66, y .

Type-locality: Singapore (A. R. Wallace).

Banguey I., N. Borneo (Emery).

A worker and two males taken in British North Borneo (E. B.
Kershaw). The male measures 8 mm., and is ferruginous red, with
the posterior part of the head and some indistinct spots on the niesono-
tum dark brown. Wings slightly infuscated, with dark brown veins
and pterostigma. Antennae very long (7 mm.), mandibles small,
narrow, edentate, with acuminate, pointed tips. Petiole 1\ times
as long as broad, narrowed in front, but with very prominent stig-
matic tubercles; in profile about as long as high, triangular, with
sloping, slightly concave anterior and abrupt posterior surface and
blunt apex. Pygidium small, bluntly rounded, cerci well-developed;
genitalia partially exserted. Body shining, sparsely and finely
punctate. Hairs brown, short, rather abundant, pubescence pale,
more abundant, and like the hairs, on all parts of the body.

19c. Diacamma rugosum subsp. vagans Smith var. birmanum

Emery.

Diacamma rugosum subsp. vagans var. birmana Emery. Ann. Mus. civ.

Genova, 1887, ser. 2, 5, p. 441, ^ ; Emery, Gen. Ins. Ponerinae, 1911, p. 67;

Forel, Rev. Suisse zool., 1911, 19, p. 23.
Diacamma rugosiwi subsp. sculptum var. birmana Emery, Rend. R. accad. sci.

Bologna, 1897, 1896-97, n. s. 1, p. 157, S .

Type-locality: Minhla, Burmah (Comotto).
Sarawak (Haviland).

wheeler: the ants of borneo. 55

20a. BoTHROPONERA iNSULARis Emery var. brevior Forel.

Pachycondyla (Bothroponera) insularis v. brevior Forel, Mitth. Naturh. mus.
Hamburg, 1901, 18, p. "45, ^ ; Emery, Gen. Ins. Ponerinae, 1911, p. 76.

Type-locality: Borneo.

21. Bothroponera rufipes (Jerdon).

Ponera rufipes Jerdon, Ann. mag. nat. hist., 1854, ser. 2, 13, p. 102, ^ .
Pachycondyla rufipes Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 106, S .
Bothroponera rufipes Mayr, Verh. Zool. bot. gesellsch. Wien, 1886, 36, p. 359;

Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 53, ^ .
Ponera {Bothroponera) rufipes Forel, Journ. Bombay nat. hist, soc, 1900, 13,

p. 323.
Pachycondyla {Bothroponera) rufipes Emery, Gen. Ins. Ponerinae, 1911, p. 76.

Type-locality: Malabar, Southern India (Jerdon).
Kapouas Basin, Borneo (Chaper).

22, Bothroponera trident ata (Smith).

Pachycondyla tridentata Smith, Cat. Hymenop. Brit. mus. 1858, 6, p. 106, ^ .
Bothroponera tridentata MajT, Ann. Mus. civ. Genova, 1872, 2, p. 149; Emery,

ibid., 1887, ser. 2, 5, p. 442, ^ .
Pachycondyla {Bothroponera) tridentata Emery, Gen. Ins. Ponerinae, 1911,

p. 77.

Type-locality: Sarawak, Borneo.
Sarawak (Doria and Beccari).

23a. Bothroponera tridentata subsp. debilior Forel.

Pachycondyla {Bothroponera) tridentata subsp. debilior Forel, Mitth. Naturh.
mus. Hamburg, 1901, 18, p. 46, ^ ; Emery, Gen. Ins. Ponerinae, 1911,

p. 77, S .

Type-locality: Tandjong, S. E. Borneo (Fritz Suck).

24. Ectomomyrmex obtusus (Emery).

Pachycondyla {Bothroponera) obtusa Emery, Ann. Mus. civ. Genova, 1900,
ser. 2, 20, p. 667 nota ^ .

56 bulletin: museum of comparative zoology.

Pachycondyla {Edomomyrmex) obtusa Emery, Gen. Ins. Ponerinae, 1911,
p. 79, y .

Type-locality: Pulo Laut, Borneo (W. Doherty).

25. EupoNERA (Brachyponera) luteipes (Mayr).

Ponera luteipes MajT, Verh. Zool. bot. gesellsch. Wien, 1862, 12, p. 722, ^ ? ;

Forel, Journ. Bombay, nat. hist, soc, 1900, 13, p. 326, ^ 9 cf .
Euponera (Brachyponera) luteipes Emery, Ann. Soc. ent. Belgique, 1901, 45,

p. 47; Emery, Gen. Ins. Ponerinae, 1911, p. 84.
Brachyponera luteipes Bingham, Fauna Brit. India. Hymenop., 1903, 2, p. 101,

y 9 cf.

Type-locality: Milu, Nicobar Islands (No vara Expedition).
Sarawak (Haviland).

26a. Euponera (Trachymesopus) darwini Forel var. indica

Emery.

Euponera (Pseudoponera) darwini var. indica Emery, Bull. Soc. ent. Ital.,

1900, 31, p. 268 nota, 9 .
Psevdoponera danvini Bingham, Fauna Brit. India. Hymenop., 1903, 2, p. 93.
Euponera (Trachymesopus) darwini var. indica Emery, Gen. Ins. Ponerinae,

1911, p. 86.

Type-locality: Upper Burmah (Doherty).
Sarawak (Haviland, Will).

27. Ponera truncata Smith.

Ponera truncata Smith, Journ. Proc. Linn. soc. London. Zool. Suppl., 1860, 4,
p. 72, 9 ; Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 150, 9 ; Emery,
Gen. Ins. Ponerinae, 1911, p. 92, 9 .

Type-locality: Celebes.
Sarawak (Doria and Beccari).

Two dealated females from Sarawak (H. W. Smith) seem to be
referable to this species, but are only 3-3.5 mm. long, whereas the
dimensions of the female cotypes are given by Mayr as 4-4.2 mm.
My specimens may represent a distinct variety or subspecies, but it
seems inadvisable to introduce another name on the basis of such
meager material.

wheeler: the ants of borneo. 57

*28. PoNERA GLEADOWi Forel.

PoneragleadotviForel, in Emery, Mem. R. acoad. sci. Bologna, 1896, 1895-1896,
ser. 5, 5, p. 292 7iota § ; Emery, ibid., p. 297, fig. 17a, b, c; Bingham,
Fauna Brit. India. H3Tnenop., 1903,2, p. 91; Emery, Gen. Ins. Poner-
inae, 1911, p. 91.

Type-localiiy: Poona, India (Wrougliton).

Two workers from Sarawak (H. W. Smith) agree very closely with
a typical specimen from Orissa received from Professor Forel. The
species has a wide distribution, occurring as far north and west as
Algeria.

*29a. PoNERA coNFiNis Roger var. javana Forel.

Ponera confinis var. javana Forel, Mitth. Naturh. mus. Hamburg, 1905, 22,
p. 6, y 9 ; Emery, Gen. Ins. Ponerinae, 1911, p. 90, ^ 9 .

Type-locality: Buitenzorg, Java (K. Kraepelin).

Seven workers and two females from Kuching (John Hewitt) agree
well with a specimen from Singapore given me by Forel, with the
description of specimens from Buitenzorg and with a worker taken
by H. W. Smith at Surubaya, Java.

30. Leptogenys (Lobopelta) chalybea Emery.

Lobopelta iridescens Mayr, Ann. Mus. civ. Geneva, 1872, 2, p. 150, ^ ; Mayr,

Verb. Zool. bot. gesellsch. Wien, 1878, 28, p. 665, ^ {nee Smith).
Lobopelta chalybea Emery, Ann. Mus. civ. Geneva, 1887, ser. 2, 5, p. 432, ^ .
Leptogenys (Lobopelta) chalybea Emery, Gen. Ins. Ponerinae, 1911, p. 102, ^ .

Type-locality: Sarawak, Borneo (Doria and Beccari).
Four workers from Kuching (John Hewitt).

31. Leptogenys (Lobopelta) iridescens (Smith).

Ponera iridescens Smith, Journ. Proc. Linn. Soc. London. Zool., 1857, 2, p. 66, U .
Lobopelta iridescens Emery, Ann. Mus. civ. Geneva, 1887, ser. 2, 5, p. 431, § .
Leptogenys (Lobopelta) iridescens Emery, Gen. Ins. Ponerinae, 1911, p. 104, ^ .

Type-locality: Sarawak, Borneo (A. R. Wallace).

Sarawak (Doria and Beccari, Haviland).

A dozen workers from Kuching (H. W. Smith).

58 bulletin: museum of comparative zoology.

31a. Leptogenys (Lobopelta) iridescens subsp. currens Forel.

Leptogenys (Lohopelta) iridescens subsp. currens Forel, Rev. Suisse zool., 1901,
9, p. 329, y ; Emery, Gen. Ins. Ponerinae, 1911, p. 104, ^ .

Type-locality: Sarawak, Borneo (Haviland).

32. Leptogenys (Lobopelta) mutabilis (Smith).

Ponera mutabilis Smith, Journ. Proc. Linn. Soc. London. Zool., 1861, 6, p. 45, ^ .
Lobopelta mutabilis Mayr, Tijdschr. ent. 1867, 10, p. 89; Emery, Ann.Mus.

civ. Genova, 1887, ser. 2, 5, p. 30, ^ .
Leptogenys {Lobopelta) mutabilis Emery, Gen. Ins. Ponerinae, 1911, p. 104.

Type-locality: Tondano, Celebes (A. R. Wallace).
Sarawak, Borneo (Doria and Beccari; Bedot and Pictet).
Eleven workers from British North Borneo (E. B. Kershaw) and
two from Kuching (H. W. Smith).

33a. Leptogenys (Lobopelta) processionalis Jerdon var. dis-

tinguenda Emery.

Lobopelta distinguemla Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5, p. 430, ^ .

Lobopelta mutabilis (part) Mayr, ibid., 1872, 2, p. 151, ^ .

Leptogenys {Lobopelta) ocellifera subsp. distinguenda Forel, Mitth. Naturh.

mus. Hamburg, 1901, 18, p. 46, ^ .
Leptogenys {Lobopelta) processionalis var. distinguenda Emery, Gen. Ins.

Ponerinae, 1911, p. 104.

Type-locality: Sarawak, Borneo (Doria and Beccari).

Tandjong, S. E. Borneo (Fritz Suck).

Two dozen workers from Kuching (H. W. Smith).

34. Leptogenys (Lobopelta) diminuta (Smith).

Ponera diminuta Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 69, ^
Lobopelta diminuta Mayr, Verh. Zool. bot. gesellsch. Wien, 1862, 12, p. 734, ^
Leptogenys diminuta Forel, Journ. Bombay nat. hist, soc, 1900, 13, p. 312.
Leptogenys {Lobopelta) diminuta Emery, Gen. Ins. Ponerinae, 1911, p. 103.

Type-locality: Sarawak, Borneo (A. R. Wallace).
S. E. Borneo (Fritz Suck).

wheeler: the ants of borneo. 59

34a. Leptogenys (Lobopelta) diminuta var. laeviceps (Smith).

Ponera laeinceps Smith, Joiirn. Proe. Linn. soc. London. Zool., 1857, 2, p. 69, ^ .

Ponera simillima Smith, ibid., 1861, 5, p. 105, ^ .

Lohopelta diminuta var. laeviceps Emery, Ann. Mus. civ. Geneva, 1887, ser. 2,

5, p. 433, S .
Leptogenys (Lobopelta) diminuta var. laeviceps Emery, Gen. Ins. Ponerinae,

1911, p. 103.

Type-localUy: Sarawak, Borneo (A. R. Wallace).
Two workers from Kuehing (John Hewitt).

*35. Leptogenys (Lobopelta) borneensis, sp. nov.

Worker. Length: 9 mm.

Long and slender. Head about ^ longer than broad, a little broader in
front than behind, with nearly straight sides in front, rounded behind, with
rather deeply excavated occipital border. Eyes rather small, situated a
distance equal to their length from the anterior corners of the head. Mandibles
rather large, triangular, with deflected tips, distinctly concave lateral, dentate
apical and denticulate basal borders. Clypeus strongly carinate, its anterior'
border entire, projecting as a membranous, rather narrowly rounded lobe.
Antennae long and slender; scapes extending about I their length beyond the
posterior corners of the head; all the funicular joints much longer than broad;
first joint a little more than half as long as the second, second sHghtly longer
than the third. Thorax long and slender. Pronotum slightly flattened above,
longer than broad, mesonotum shorter and much narrower and lower than the
pronotum, its dorsal outline very feebly concave; epinotxrm scarcely broader
than the mesonotvim, but very distinctly higher and longer, the base straight
and horizontal, twice as long as the vertical declivity into which it passes
through a curve without any trace of an angle. The sides of the declivity are
feebly marginate. Each epinotal stigma is situated in a sharply defined
eUiptical depression. Petiole from above fully twice as long as broad, nar-
rowed in front, laterally compressed. In profile the node is longer than high,
its anterior border long and convex, the posterior border straight, the apex
blunt, the ventral surface of the petiole sinuous in the middle. Gaster small.
Legs long and slender.

Shining, very sparsely and very finely punctate; mandibles very finely and
densely striate.

Hairs whitish, short, sparse, and erect on the body, more abundant and
more oblique on the appendages. Pubescence pale, present only on the
funiculi and tarsi.

Castaneous; mandibles, clypeus, legs including coxae, neck, prosterna

60 bulletin: museum of comparative zoology.

petiole and tip of gaster, red; posterior margins of gastric segments yellowish.

Described from a single specimen taken at Kuching by Mr. John
Hewitt.

This species has much the same color as iridescens apart from the
blue reflections, but the shape of the node and thorax serve to dis-
tinguish it at once.

36a. Leptogenys (Lobopelta) kitteli Mayr subsp. laevis Mayr.

Lobopelta kitteli var. laevis Mayr, Verh. Zool. bot. gesellsch. Wien, 1878, 28,

p. 665, ^ .
Leptogenys (Lobopelta) kitteli subsp. laevis Forel, Rev. Suisse zool., 1901, 9,

p. 329, y ; Emery, Gen. Ins. Ponerinae, 1911, p. 104.

Type-locality: Java.
Sarawak (Haviland).

37. Odontomachus haematoda (Linne).

Formica haematoda Linne, Syst. nat., ed. 10, 1758, p. 582, ^ .

Myrmecia unispinosa Fabricius, Syst. Piez., 1804, p. 423, ^ .

Myrmecia haematoda Fabricius, Syst. Piez., 1804, p. 423, ^ .

Odontomachus haematodes Latreille, Hist. nat. Crust. Ins., 1805, 13, p. 257.

Ponera {Odontomachiis) haematoda Latreille, ibid., 1809, 4, p. 128, ^ .

Odontomachus haematoda Dalla Torre, Cat. Hymenop., 1893, 7, p. 50; Emery,

Gen. Ins. Ponerinae, 1911, p. 114, pi. 3, fig. 18, ^ 9 &.
Formica maxillosa DeGeer, Mem. hist, ins., 1773, 3, p. 601, ^ , pi. 31, fig. 3-5.
Formica unispinosa Fabricius, Ent. syst., 1793, 2, p. 359, S .
Odontomachus simillimvs Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 80,

pi. 5, fig. 8, 9.

Type-locality: South America (Rolander).

Sarawak (Doria and Beccari), Kapouas Basin (diaper).

Two workers from Kuching (John Hewitt).

38. Odontomachus rixosus Smith.

Odontomachus rixosus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 67, ^ ; Emery, Gen. Ins. Ponerinae, 1911, p. 114.

Type-locality: Singapore (A. R. Wallace).

wheeler: the ants of Borneo. 61

Sarawak (J. Doria and O. Beccari); Kapouas Basin (Chaper).
A dozen workers from Kuching (H. W. Smith).

39. Odontomachus malignus Smith.

Odontomachus malignus Smith, Journ. Proc. Linn. soc. London. Zool., 1859,
3, p. 144, ^ ; Emery, Gen. Ins. Ponerinae, 1911, p. 113.

Odontomachus tuherculatus Roger, Berl. ent. zeitschr., 1861, 5, p. 28, y ; Mayr,
Ann. Mus. civ. Genova, 1872, 2, p. 149, ^ .

Type-locality: Aru {k. R. Wallace).
Sarawak, Borneo (Doria and Beccari).

40. Anochetus agilis Emery.

Anochetus agilis Emery, Ann. Soc. ent. Belgique, 1901, 45, p. 53, ^ ; Emery,
Gen. Ins. Ponerinae, 1911, p. 108.

Type-locality: Banguey, Borneo (Staiidinger and Bang-Haas).

Dorylinae.
41. DoRYLUS (Dichthadia) laevigatus (Smith).

TtjphlopoJie laevigatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 70, ^ .
Dichthadia glaherrima Gerstacker, Stettin, ent. zeit., 1863, 24, p. 93, 9 .
Dorylus klugi Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5, p. 448, pi. 1,

fig. 10, cf.
Dorylus laevigatus Emery, Zool. jahrb. Syst., 1895, 8, p. 729, S .
Dorylus (Dichthadia) laevigatus Emery, Gen. Ins. Dorylinae, 1910, p. 8, y 9 cT.

Type-locality: Sarawak, Borneo (A. R. "Wallace).

Sarawak (Doria and Beccari).

Two large workers from Kuching (John Hewitt).

42. Aenictus laeviceps (Smith).

Typhlatta laeviceps Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 79, y .
Aenictus laeviceps Forel, Ann. Soc. ent. Belgique, 1890, 34, C. R., p. Oil, ^ ;

Emery, Gen. Ins. Dorylinae, 1910, p. 30, S .

)

62 bulletin: museum of comparative zoology.

Type-hcality: Sarawak, Borneo (A. R. Wallace).

Sarawak (Doria and Beccari).

Five workers from Bongo Mt., Borneo (Hewitt and Brooks).

43. Aenictus gracilis Emery.

AenicttLS gracilis Emery, Rev. Suisse zool., 1893, 1, p. 187, ^ ; Emery, Gen.
Ins. Dorylinae, 1910, p. 30, S .

Type-locality: Sarawak (Bedot and Pictet).

44. Aenictus cornutus Forel.

Aenictus cornutus Forel, Ann. Soc. ent. Belgique, 1900, 44, p. 75, ^ ; Emery,
Gen. Ins. Dorylinae, 1910, p. 29, g .

Type-locality: Sarawak (Haviland).

A single worker from Kuching (John Hewitt).

45. Aenictus punctiventris Emery.

Aenictus punctiventris Emery, Bull. Soc. ent. Ital., 1901, 33, p. 47, cf ; Emery,
Gen. Ins. Dorylinae, 1910, p. 31, d^.

Type-locality: Borneo.

A single worker from Kuching (John Hewitt).

46. Aenictus aitkeni Forel.

Aenictus aitkeni Forel, Journ. Bombay nat. hist. soc. 1901, 13, p. 46.5, 475, y ;
Bingham, Fauna Brit. India. Hymenop., 1903, 2, p. 19, fig. 18, ^ ; Emery,
Gen. Ins. Dorylinae, 1910, p. 29, ^ .

Type-locality: Kanara, India (Aitken).
Sixteen workers from Kuching (H. W. Smith).

Myrmicinae.

*47. Metapone hewitti, sp. nov.

Male. Length 6-7 mm.

Body long and slender. Head as broad as long, evenly convex and rounded
behind, without posterior corners; cheeks very short; eyes moderately large.

wheeler: the ants of borneo. 63

but not very convex; ocelli rather small. Mandibles small, but well-developed,
their external borders slightly sinuate towards the base, convex at the tips;
apical and basal borders distinct, subequal, the former with four subequal
teeth. Clypeus large, convex, somewhat broader than long, slightly depressed
or flattened posteriorly. Front truncated anteriorly, with a transverse crest
or carina separating it from the preocellar space and connecting the frontal
carinae, which are prominent, nearly straight, subparallel and as far apart
as they are from the lateral borders of the head. Posteriorly each carina
curves forward medially to the eye as a distinct ridge and terminates opposite
its anterior end, thus enclosing a small, shallow, elliptical scrobe about the
base of the antenna. Antennae 12-jointed, scape very small, about twioe
as long as broad, first funicular joint also very small, broader than long, sub-
globular; second joint longer but also broader than long, the remaining nine
joints cylindrical, of equal breadth, distinctly longer than broad and gradu-
ally increasing in length distally; terminal joint nearly as long as the two
preceding joints together, with tapering and somewhat pointed tip. Thorax
long, narrower than the head through the eyes. Pronotima well-developed,
truncated in front; mesonotum and scutellum somewhat flattened above,
the former with distinct Mayrian furrows, the latter with a peculiar blunt,
spatulate spine on each side, slightly curved inward at its tip. Epinotum
longer than broad, subrectangular from above, its base horizontal and twice
as long as the vertical decUvity into which it passes through an abrupt curve,
the sides of the declivity above and of the base coarsely and rather irregu-
larly marginate. Petiole with a short, stout peduncle anteriorly and a thick,
cuboidal node, which is a little longer than broad and slightly higher in
front than behind, with tnmcated anterior and posterior and feebly rounded
dorsal and lateral surfaces. Seen in profile its ventral margin is slightly
bisinuate, with a small, triangular tooth at the anterior end of the peduncle.
Postpetiole distinctly broader than the petiole and broader than long, from
above transversely elliptical, in profile slightly truncated anteriorly, convex
and rounded above, its ventral border unarmed, nearly straight. Gaster
elongate elliptical, with straight anterior border and tapering tip. Genitalia
completely retracted, cerci apparently absent; pygidium and hypopygium
short and pointed. Legs short, of the usual simple form, without the tibial
spines of the worker and female; spurs of the middle and hind tibiae simple,
blunt at the tip. larsal claws very small, strongly curved, nonpectinated.
Wings very short (4.5 mm.), with a well-developed discoidal cell, a single
cubital cell and the radial cell slightly open at the tip. The radial cell is
large. Pterostigma well-developed and conspicuous.

Subopaque; mandibles opaque, longitudinally rugose and very finely
punctate. Head reticulate-rugose, the clypeus more coarsely and trans-
versely. Front behind its anterior truncation with regular longitudinal rugae
converging to the anterior ocellus. Antennal scrobes less distinctly longi-
tudinally rugose. Upper surface of mesonotum and scutellum and sides of
thorax sharply and regularly longitudinally rugose, with elongate, shallow

64 bulletin: museum of comparative zoology.

foveolae in the narrow interrugal spaces on the mesonotum. Dorsal surface
of epinotum, including the upper portion of the decUvity, with extremely
coarse reticulate rugae, some of which are clearly transverse. Petiole above
less coarsely and even more irregularly rugose. Postpetiole and gaster very
finely and densely punctate, with superimposed small, sparse, and very regular
piligerous punctures.

Hairs grayish brown, short, rather abundant, erect on the head, thorax, and
petiole, mostly subappressed or oblique on the postpetiole, gaster, and legs.
Antennal funicuh with very short, fine hairs, or pubescence. Wings minutely
hairy.

Black; mandibles, antennae, legs, and tip of gaster reddish brown, the tarsi
slightly paler. Wings grayish hyaline, with slightly infuscated tips and
anterior margin; veins sharply defined, brown; pterostigma dark brown.

Described from four specimens taken by Mr. John Hewitt at
Kuching in 1908. Type.— M. C. Z. 8,946.

I have described this male in detail and given it a name, although
in closely resembles the male of M. greeni Forel from Ceylon, described
from a mature pupa, because no adult winged male of the genus has
been described. The Bornean specimens may belong to a different
species, possibly M. sauicri Forel of Formosa or M. bakeri Wheeler
of the Philippines, both known only from females. It can hardly
be the male of the only other known species of Metapone, M. mjobcrgi
Forel of Queensland. The four specimens of M. hewitti have been
in my collection for many years and were placed provisionally with
Cataulacus. Forel's very careful description and figures (Rev. Suisse
zooL, 1911, 19, pi. 14) finally enabled me to recognize them as Meta-
pone males. Forel is, I believe, in error in stating that the antennae
of the male M. greeni are 11-jointed. He has apparently overlooked
the second funicular joint. As Green has shown, the species of this
extraordinary genus nest in decayed branches. He found the types
of M. greeni and their larvae in company with termites.

48. Tetraponera nigra (Jerdon).

Edlon nigrum Jerdon, Madras journ. lit. sci., 1851, 17, p. 112, § .
Tetraponera atrata Smith, Ann. mag. nat. hist., 1852, ser. 2, 9, p. 44, ^ ;

Smith, Journ. Proc. Linn. soc. London. ZooL, 1857, 2, p. 70, 9 .
Pseudoynyrma nigra Smith, Cat. Hymenop., Brit, mus., 1858, 6, p. 159, ^ .
Pfendomyrma airata Smith, ibid. p. 159.
Pseudomyrina carbonaria Smith, Journ. Proc. Linn. soc. London ZooL, 1863,

7, p. 20, y 9 .

Sima nigra Emery in Dalla Torre, Cat. Hymenop., 1893, 7, p. 54.

wheeler: the ants of borneo. 65

Type-locality: Malabar, Southern India (Jerdon).
Sarawak (A. R. Wallace).

49. Tetraponera attenuata Smith.

Tetraponera attenimta Smith, Trans. Ent. soc. London, 1877, p. 71, ^ .
Sinia attenuata Ern. Andre, Mem. Soc.'zool. France, 1892, 5, p. 53, § .
Sima (Tetraponera) attenuata var. tenuissima Emery, Ann. Mus. civ. Geneva,
' 1900, ser. 2, 20, p. 675, ^ 9 , fig. 5b.

Type-locality: Sarawak.

Kapouas Basin (Chaper); Tandjong, S. E. Borneo (Fritz Suck).

A single worker from Kuching (John Hewitt).

*50. Tetraponera difficilis Emery.

Sima {Tetraponera) difficilis Emery, Ann. Mus. civ. Geneva, 1900, ser. 2, 20,
p. 676, ^ .

Type-locality: Benculen, Sumatra (E. Modigliani) .
Six workers from Kuching (John Hewitt).

51. Tetraponera pilosa (Smith).

Pseudoponera pilosa Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 160, ^ .
Sima pilosa Emery, Ann. Mus. civ. Geneva, 1887, ser. 2, 5, p. 449 7iota; Emery,
ibid., 1900, ser. 2, 20, p. 675.

Type-locality: Borneo.

Two workers and a dealated female from Kuching (John Hewitt).

52. Myrmica ritae Emery.

Myrmica ritae Emery, Ann. Mus. civ. Geneva, 1889, ser. 2, 7, p. 501, ^ , pi. 11,
fig. 27; ibid., 1894, ser. 2, 14, p. 451, g .

Type-locality: Mt. Moolej-it (1,000-1,900 m.), Tenasserim (L. Fea).
Pulo Laut, Borneo (W. Doherty).

*53. Pheidole megacephala (Fabricius).

Formica megacephala Fabricius, Ent. syst., 1793, 2, p. 361, Ql.
Formica {Myrmica) trinodis Lesana, Mem. Accad. sci. Turino, 1834, 37, p.
327, pi. 36, fig. 6.

66 bulletin: museum of comparative zoology.

Oecophthora pusilla Heer, Hausameise Madeiras, 1852, p. 15, Ql ^ 9 o"^, pi. 1,

fig. 1-4.
Myrmica? laevigata Smith, Trans. Ent. soc. London, 1855, ser. 2, 3, p. 130, ^ ,

pi. 9, fig. 7, 8.
Myrmica trinodis Mayr. Verb. Zool. bot. gesellsch. Wien, 1855, 5, p. 414,

noia, S .
Myrmica (Pheidole) laevigata Smith, Cat. Brit, fossor. H3Tiienop., 1858, p. 35,

225, y .
Pheidole pusilla Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 173, pi. 9, fig.

18-20.
Pheidole jamts Smith, ibid., p. 175, 9 , pi. 9, fig. 13-17.
Pheidole megacephala Roger, Verzeich. formicid., 1863, p. 30; Dalla Torre,

Cat. Hymenop., 1893, 7, p. 92.

Type-locality: Mauritius (Coll. Bosc).

Numerous soldiers and workers from Kuching (John Hewitt and
H. W. Smith).

54. Pheidole javana Mayr.

Pheidole javana Mayr, Tijdschr. ent., 1867, 10, p. 98, 21 ^ ,

Type-locality: Java.

Sarawak (Doria and Beccari; Bedot and Pictet); Kapouas Basin
(Chaper).

Two soldiers and several workers from Kuching (John Hewitt) and
Sarawak (H. W. Smith).

*55. Pheidole bugi, sp. nov.

Soldier. Length 2 mm.

Head large, subrectangular, longer than broad, with nearly straight sub-
parallel sides, slightly narrowed at the posterior corners, with deeply and
angularly excised occipital border, distinctly depressed in the occipital region
and with distinct occipital groove. Eyes small, with angular inferior orbits,
situated at the anterior fourth of the head. Mandibles large, convex, with
broad apical margins, furnished at their tips with a pair of large, blunt teeth.
Clypeus flattened, or slightly concave in the, middle, ecarinate, its anterior
border deeply notched. Frontal area distinct, semicircular, impressed.
Frontal carinae rather short, diverging behind and bordering flattened,
indistinct scrobe-like areas. Antennae short and slender, the scapes reaching
the lateral borders of the head a little behind the eyes and in front of the middle;

wheeler: the ants of Borneo. 67

joints 2-7 of the funiculi distinctly broader than long; club somewhat longer
than the remainder of the funiculus. Thorax robust, the pronotum broad
and convex, its humeri protruding as bluntly rounded angles; mesonotum
sloping, with a transverse torus in the middle. Epinotum low, its base in
profile straight and longer than the declivity; the spines short, rather erect,
much shorter than the base of the epinotum and shorter than their distance
apart at their insertions. Petiole short, the anterior slope of the node very
concave, its summit blunt, transverse, and rather deeply emarginate, its
posterior slope abrupt. Postpetiole slightly broader than the petiole, trans-
verse, very convex above, broader in front than behind, the sides rounded.
Gaster broadly elliptical, flattened, smaller than the head, with straight
anterior border. Legs with stout, slightly swollen femora.

Somewhat shining; mandibles smooth, minutely and sparsely punctate.
Clypeus smooth and shining in the middle, rugose on the sides; remainder of
head sculptured, the anterior f longitudinally rugose, with feebly reticulate
interrugal areas, especially on the sides, the posterior third reticulately rugose,
the scrobe-like areas densely punctate. Neck, pronotum, and mesonotum
more finely reticulate-rugose; epinotum smooth and shining; petiole and
postpetiole subopaque, indistinctly and very finely punctate or alutaceous.
Gaster and legs smooth and shining, sparsely and finely punctate.

Hairs yellow, erect or suberect, coarse, rather long, of uneven length, more
abundant on the body than on the appendages.

Ferruginous; gaster darker, brown; legs and antennae paler, more j^ellow-
ish; borders of mandibles and clj^peus blackish.

Worker. Length 1.4 nun.

Head 3- longer than broad, subrectangular, with very feebly convex sides
and feebly sinuate posterior border, as broad in front as behind. Eyes very
small, at the middle of the sides of the head. Mandibles with oblique apical
margins furnished with four acute, subequal teeth. Clypeus short, convex,
its anterior border broadly rounded, entire. Antennal scapes reaching the
posterior border of the head. Thorax rather slender; pro- and mesonotum
fused, feebly rounded above and on the sides; mesoepinotal constriction short
and deep. Epinotum with subeqvial base and declivity, the spines reduced
to small, rather blunt teeth, not longer than broad at their bases. Superior
border of petiolar node transverse and entire. Postpetiole much as in the
soldier.

Smooth and shining; mesopleurae and sides of epinotum densely punctate;
petiolar and postpetiolar nodes subopaque.

Hairs whitish, erect, more uniform and somewhat more abundant than in
the soldier, especially on the legs and scapes.

Yellowish brown; head and gaster a little darker; thorax and appendages
paler and more yellowish.

68 bulletin: museum of compakative zoology.

Described from a single soldier and four workers from Sarawak,
(Roland Thaxter). Type— M. C. Z. 8,947.

This species is evidently related to Ph. nodgii Forel of Java, but
the soldier is smaller, with more deeply notched clypeiis, much less
distinct antennal scrobes, shorter epinotal spines, more deeply notched
petiolar node, laterally less angular postpetiole, and very different
thoracic sculpture. The worker bugi is also smaller than that of
nodgii, has a more elongate head, very feebly armed epinotum, a
more rounded postpetiole, and very different sculpture.

56. Pheidole aristotelis Forel.
Pheidole aristotelis Forel, Rev. Suisse zool., 1911, 19, p. 43, % ^ d'.
Type-locality: Sarawak, Borneo (Haviland).

57. Pheidole comata Smith.

Pheidole comata Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 176, ^ ; Mayr,
Verb. Zool. hot. gesellsch. Wien, 1886, 36, p. 360.

Type-locality: Sarawak, Borneo.

58. Pheidole havilandi Forel.

Pheidole havilandi Forel, Rev. Suisse zool., 1911, 19, p. 38, Ql S 9 cT.
Type-locality: Sarawak, Borneo (Haviland).

59a. Pheidole sauberi Forel subsp. sarawakana Forel.
Pheidole sauberi subsp. sarawakana Forel, Rev. Suisse zool., 1911, 19, p. 45,

Type-locality: Sarawak, Borneo (Haviland).

60. Ischnomyrmex longipes (Smith).

Myrmica longipes Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 70,

y , pi. 1, fig. 6.
Myrmica (Monomorium) longipes Smith, Cat. Hymenop. Brit, mus., 1858, 6,

p. 126, y .

wheeler: the ants of Borneo. 69

Ischnomyrmex longipes Mayr, Tijdschr. ent., 1867, 10, p. 60, ^ ; Ern. Andre,

Mem. Soc. zool. France, 1892, 5, p. 53, ^ .
Aphaenogaster longipes Emery, Ann. Mus. civ. Geneva, 1888, ser. 2, 5, p. 531,

S , pi. 9, fig. 2.
Aphaenogaster (Ischnomyrmex) longipes Forel, Rev. Suisse zool., 1911, 19,

p. 24, y .
Pheidole (J sopheidole) longipes Forel, Rev. Suisse zool., 1912, 20, p. 765, 21 ^ .
Pheidole {Ischnomyrmex) longipes Forel, Zool. jahrb. Syst., 1913, 36, p. 49,

21 9, fig. N.

Type-locality: Sarawak, Borneo {X. R. Wallace).
Sarawak (Haviland); Kapouas Basin (Chaper).
Nine workers from the Rambungan River, Sarawak (H. W. Smith)
and one from Kuching (John Hewatt).

61. Myrmicaria carinata (Smith).

Heptacondylus carinatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 73, ^ .
?Physatta dromedarius Smith, ibid., p. 78, 9 .
Myrmicaria carinata Dalla Torre, Cat. Hymenop., 1893, 7, p. 155.
Myrmicaria fodiens race carinata Emery, Rev. Suisse zool., 1893, 1, p. 219.

Type-locality: Sarawak, Borneo (A. R. Wallace).

Tandjong, S. E. Borneo (Fritz Suck).

Two workers and two males from Kina Balu, N. Borneo, obtained
from Staudinger. The worker has the gaster distinctly striated at the
base. This character, not mentioned by Smith, nevertheless exists
in the type {teste W. F. Kirby) and is regarded by Emery as distinctive
of the species.

*61a. Myrmicaria carinata subsp. gagatina, subsp. nov.

Worker. Length 5.7 mm.

The series of small angles formed by the lateral carmae of the meso- and
epinotum are somewhat more acute than in the topical form and the body is
much smoother and more shining. There are only a few delicate longitudinal
rugae on the head, some confined to the sides and just below and above the
eyes and some abbreviated and widely separated on the posterior portion of
the head. On the thorax the rugae are also finer, more regular and further
apart. The extreme base of the gaster is finely striated as in typical carinata.

70 bulletin: museum of comparative zoology.

The color, however, is very different, the body being jet black, with the mandi-
bles, antennae, legs, neck, and articulations of the pedicel dark reddish brown.
The hairs covering the body and appendages are very dark brown, almost
black.

Described from a single worker taken by Mr. G. E. Bryant on
Matang Mt., West Sarawak and sent me by Mr. Horace Donisthorpe.

62. Myrmicaria subcarinata (Smith).

Heptocondylus subcarinatus Smith, Journ. Proc. Linn. soc. London. Zool.,

1857, 2, p. 73, S .
Physatta gibbosa Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 172, 9 .
Heptacondylus dromedarivs Mayr, Verh. Zool. hot. gesellsch. Wien., 1862, 12,

p. 757, y (nee Smith).
Myrmicaria subcarinata Mayr, Tijdschr. ent., 1867, 10, p. 112, y 9.
Myrmicaria fodiens race subcarinata Emery, Rev. Suisse zool., 1893, 1, p. 219.

Typc-localiti/: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari ; Haviland) ; Kapouas Basin (Chaper) ;
Tandjong (Fritz Suck).

Two workers from British North Borneo (E. B. Kershaw).

63. Myrmicaria rugosa (Smith).

Heptacondylus rugosus Smith, Journ. Proc. Linn. soc. London. Zool. Suppl.,

1860, 4, 110, y .
Myrmicaria (Heptacondylus) rugosus Smith, ibid., 1864, 8, p. 73, ^ 9 cf .
Myrmicaria rugosa Mayr, Tijdschr. ent. 1867, 10, p. 113, ^ .

Type-locality: Batjan (A. R. Wallace).

Tandjong, S. E. Borneo (Fritz Suck). ,

64. Myrmicaria arachnoides (Smith).

Heptacondylus arachnoides Smith, Journ. Proc. Linn. soc. London. Zool., 1857,

2, p. 72, y 9 .
Heptacondylus longipes Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 142, ^ .
Myrmicaria longipes Mayr, Tijdschr. ent., 1867, 10, p. 113, S 9 c?.
Myrmicaria arachnoides Emery, in Dalla Torre, Cat. Hymenop., 1893, 7, p. 155.

Type-locality: Sarawak, Borneo (A. R. Wallace).

wheeler: the ants of borneo. 71

Tandjong, S. E. Borneo (Fritz Suck); Kapouas Basin (Chaper);
Sarawak (Doria and Beccari).

A male, female, and worker from Kucliing (John Hewitt) and a male
and a dozen workers from the same locality (H. W. Smith). Hewitt's
specimens are accompanied by two of the peculiar nests of this ant,
which consist of coarse, fibrous carton, built in the form of a series
of contiguous and rather irregular chambers on the under surfaces of
large leaves. One of the nests is 9 cm. long, 5 cm. broad and 2-3 cm.
high, the other 8 cm. long, 4 cm. broad and 2 cm. high. These nests
have been observed by Jacobson and von Buttel Reepen in Java and
are briefly described by Forel (Notes of the Leyden mus., 1909, 31,
p. 252 and Zool. jahrb. Syst, 1913, 36, p. 73).

*64a. Myrmicaria arachnoides subsp. melanogaster Emery.

Emery, Ann. Mus. civ. Geneva, 1900, ser. 2, 20, p. 692 nota, ^ .
Forel, Rev. Suisse zool., 1911, 19, p. 23, ^ 9 c?.

Type-locality: Sarawak, Borneo.
Sarawak (Haviland); Hayvep (Zimmer).

*65. Cardiocondyla nuda (Mayr).

Leptothorax nudus Mayr, Sitzb. Akad. wiss. Wien, 1866, 53, p. 508, S .
Cardiocondyla nuda Forel, Mitth. Mlinch. ent. ver. 1881, 5, p. 3, ^ .

Type-locality: Ovalau, Fiji (Mus. Godeffroy).
A single worker from Sarawak (H. W. Smith).

66. Crematogaster brunnea Smith,

Crernatogaster brunneus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 75, ^
Cremastogaster brunnea Mayr, Verh. Zool. hot. gesellsch. Wien, 1886, 36,

p. 360, y .
Cremastogaster brunea Dalla Torre, Cat. Hymenop., 1893, 7, p. 80.

Type-locality: Sarawak, Borneo (A. R. Wallace).
A single worker from Sarawak (H. W. Smith) agrees well with
Smith's description of the worker minor of this species although the

72 bulletin: museum of comparative zoology.

color is somewhat darker. Unfortunately the thorax is somewhat
crushed so that its precise form cannot be determined.

67. Crematogaster cephalotes Smith.

Crematogaster cephalotes Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 75, y {nee Gerstaecker) .
Cremasicgaster cephalotes Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 53, ^ .

Type-locality: Sarawak, Borneo (A. R. Wallace).
Kapouas Basin (Chaper).

68. Crematogaster coriaria Mayr.

Cremastogaster coriaria Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 154, y ;
Emery, ibid., 1887, ser. 2, 5, p. 467, ^ .

Type-locality: Sarawak, Borneo (Doria and Beccari).

69a. Crematogaster egidyi Forel subsp. spinozae Forel.
Cremastogaster egidyi subsp. spinozai Forel, Rev. Suisse zool., 1911, 19, p. 25,

y 9.

Type-locality: Sarawak, Borneo (Ha\aland).

69b. Crematogaster egidyi subsp. spinozae var. hayvepana Forel.

Cremastogaster egidyi subsp. spinozai var. hayvepana Forel, Rev. Suisse zool.,
1911, 19, p. 26, y 9 .

Type-locality: Hayvep, Borneo (Winkler).

70. Crematogaster ferrarii Emery.

Cremastogaster ferrarii Emery, Ann. Mus. civ. Genova, 1888, ser. 2, 5, p. 533,
y ; Emery, Rev. Suisse zool., 1893, 1, p. 193.

Type-locality: Siboga, Smnatra (E. Modigliani).
Sarawak, Borneo (Bedot and Pictet).

7L Crematogaster fraxatrix Forel.

Cremastogaster fraxatrix Forel, Rev. Suisse zool., 1911, 19, p. 28, ^ .
Type-locality: Sarawak, Borneo (Haviland).

wheeler: the ants of borneo. 73

72. Crematogaster innocens Forel.

Cremastogaster innocens Forel, Rev. Suisse zool., 1911, 19, p. 30, ^ 9.
Type-locality: Hayvep, Borneo (Winkler).

73. Crematogaster longipilosa Forel.

Cremastogaster lovgipilosa Forel, Ann. Mus. nat. Hungar., 1907, 5, p. 24, ^ ;
Forel, Rev. Suisse zool., 1911, 19, p. 24, ^ .

Typc-localiiy: Kwala Lampur, Malacca (Biro).

Sarawak, Borneo (Haviland).

Eleven workers from Sarawak (H. W. Smith; Roland Thaxter).

74. Crematogaster modiglianii Emery.

Crematogaster modiglianii Emery, Ann. Mus. civ. Geneva, 1900, ser. 2, 20,
p. 688, S .

Type-localiiy: Sipora, Mentawei (E. IModigliani).
Banguey, Borneo (Coll. Emery).

74a. Crematogaster modiglianii var. sarawakana Forel.

Cremastogaster modiglianii v. sarawakana Forel, Rev. Suisse zool., 1911, 19,
p. 25, y .

Type-locality: Sarawak, Borneo (Haviland).

Ten workers from the Rambungan River, Sarawak (H. W. Smith).

75. Crematogaster myops Forel.

Cremastogaster myops Forel, Rev. Suisse zool., 1911, 19, p. 31, ^ 9 .
Type-locality: Sarawak, Borneo (Haviland).

76. Crematogaster obscura Smith.

Crematogaster obscnra Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 76, y ; Smith, Cat. Hymenop. Brit, mus., 18.58, 6, p. 137, ^ .

Type-locality: Sarawak, Borneo (A. R. Wallace).

74 bulletin: museum of comparative zoology.

77a. Crematogaster rogenhoferi Mayr var. fictrix Forel.

Cremastogasier rogenhoferi var. fictrix Forel, Rev. Suisse zool., 191 1, 19, p. 27, y .
Type-locality: Sarawak, Borneo (Haviland).

78. Crematogaster subcircularis Mayr.

Cremastogasier anthracina Mayr, Tijdschr. ent., 1867, 10, p. 75, ^ ; Mayr,

Ann. Mus. civ. Geneva, 1872, 2, p. 154 {nee Smith).
Cremastogasier subcircularis MajT, Verb. Zool. bot. gesellsch. Wien, 1878, 28,

p. 681, 685, y ; Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5, p. 467, ^ .

Type-locality: Borneo (Doria and Beccari).

79. Crematogaster subnuda Mayr.

Cremasiogaster svbnuda Mayr, Verb. Zool. bot. gesellsch. Wien, 1878, 28, p. 680,
682, y ; Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 53, ^ .

Type-locality: Calcutta, India.
Kapouas Basin, Borneo (diaper).

80. Crematogaster (Physocrema) diformis Smith.

Crematogaster difformis Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 76, S .
Crematogaster ampuUaris Smith, ibid., 1861, 6, p. 47, ^ .
Cremastogaster difformis Mayr, Tijdschr. ent., 1867, 10, p. 75, ^ .
Cremastogaster edentata Mayr, ibid., p. 104, 9 , pi. 2, fig. 10.
Cremastogaster deformis Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 5,

p. 467, y.
Crematogaster deformis Emery, Rev. Suisse zool., 1893, 1, p. 193.

Type-locality: Singapore (A. R. Wallace).

Sarawak (Doria and Beccari; Bedot and Pictet); Kapouas Basin
(Chaper).
A single worker from Kuching (John Hewitt).

wheeler: the ants of borneo. 75

*80a. Crematogaster (Physocrema) diformis subsp. physo-

THORAX Emery.

Cremastogastcr difformis st. physothorax Emery, Ann. Mus. civ. Geneva, 1889,

ser. 2, 7, p. 506, ^ .
Cremasiogaster physothorax Dalla Torre, Cat. Hymenop., 1893, 7, p. 84.

Type-locality: Thagata, Tenasserim (L. Fea).
Four workers from Kuching (John Hewitt).

80b. Crematogaster (Physocrema) diformis subsp. sewardi

Forel.

Cremastogaster deformis subsp. sewardi Forel, Mitth. Naturh. mus. Hamburg,
1901, 18, p. 64, S 9.

Type-locality: Borneo (Seward).

81. Crematogaster (Physocrema) inflata Smith.

Crematogaster inflatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 76, U , pi. 2, fig. 2.
Cremastogaster inflata Mayr, Tijdschr. ent., 1867, 10, p. 74, ^ ; Mayr, Ann.

Mus. civ. Geneva, 1872, 2, p. 153; Emery, ibid., 1887, ser. 2, 5, p. 466, y .

Type-locality: Singapore (A. R. Wallace).
Sarawak (Doria and Beccari).

Several workers from Kuching (John Hewitt; H. W. Smith) and
one from the Rambungan River, Sarawak (H. W. Smith).

*82. Crematogaster (Physocrema) stethogompha, sp. nov.

Worker. Length 3-6 mm.

Head slightly broader than long, slightly broader behind than in front, with
rounded, convex sides and broadly concave posterior border. Mandibles
stout, rather convex, with obhque, coarsely 5-toothed apical borders. Clyp-
eus moderately convex, ecarinate, with strongly depressed, straight, anteriorj
border. Eyes- small, flattened, just behind the middle of the head. Fronta
carinae short, subparallel; frontal area small, triangular; frontal groove short
and indistinct. Antennae 11-jointed; scapes reaching the posterior corners of

76 bulletin: museum of comparative zoology.

the head; funiculi with 4-jointed club, all the joints longer than broad. Tho-
rax resembling that of C. deformis, but the epinotum is less swollen above and
is armed with two sharp, backwardly directed spines, which are longer than
broad at their distinctly swollen bases. Promesonotal suture absent, pro- and
mesonotum together, excluding the neck, as long as broad, robust, only slightly
narrowed behind to the mesoepinotal suture, which is distinct and transverse,
but not impressed. The epinotum is broader than long, broader than the pro-
and mesonotum, bulging on the sides and dorsally in front so that the base is
very convex in profile. The declivity falls off vertically between the spines; it
is as long as the base and longitudinally grooved in the middle. A more or less
distinct carina or prominent ruga runs along the middorsal line of the thorax
from the neck to the epinotal declivity, with an interruption at the meso-
epinotal suture. Petiole longer than broad, as broad in front as behind, octag-
onal, the anterior and posterior borders longer than the six other sides, which
are subequal. In some specimens, especially in the small workers, the angles
become rounded so that the petiole may be described as elongate elliptical.
Its upper surface is flat; in profile it is slightly thicker at the posterior than at
the anterior end. Postpetiole as broad as the petiole, contracted posteriorly,
its anterodorsal surface flat, bilobed behind, but without a median longitudinal
furrow in large specimens, feebly sulcate in small specimens. Gaster thick
and broad, triangular, with straight anterior border, rounded anterior corners
and pointed tip. Legs moderately long and stout.

Opaque; gaster and mandibles distinctly shining; the latter coarsely
longitudinally striate and punctate. Clypeus and head densely punctate and
reticulate-rugose, with more prominent longitudinal rugae anteriorly, growing
finer and merging with the punctate surface on the occiput, sides, and posterior
corners. Thorax densely punctate, the upper surface of the pro- and mesono-
tum and base of the epinotum also vermiculately rugose. Petiole, postpetiole,
and first gastric segment much more finely and densely punctate than the
thorax, the first segment also with sparser and somewhat larger, piligerous
punctures; remaining gastric segments, legs, and scapes smoother, more
shining, transversely alutaceous; the legs with sparse, coarse piligerous
punctures.

Hairs short, whitish, delicate, sparse, and erect, of uneven length on the
body, more numerous on the head and thorax than on the pedicel and gaster;
shorter and more appressed on the legs and antennae. Pubescence pale,
rather long, sparse, very distinct on the gaster and sides and posterior corners
of the head.

Dark brown; mandibles, except the teeth, gula, and sometimes the lateral
and posterior corners of the head, sides of epinotum, upper surface of petiole
and articulations of legs, ferruginous red, second to fourth tarsal joints reddish
yellow.

Numerous workers taken by Harrison W. Smith near Kuching
(type-locality) and three workers from the Ranibungan River, Sara-
wak. 2\jpi.— M. C. Z. 8,948.

avheeler: the ants of borneo. 77

This species is quite distinct from the other East Indian Physo-
cremas, especially dcformis, inflata, and tumichda Emery, and easily
recognized by its large size, coarse sculpture, and well-developed
epinotal spines. It is much more closely related to a form described
by Forel from Malacca as C. dcformis subsp. vacca (Bull. Soc. Vaud.
sci. nat., 1911, ser. 5, 47, p. 384) but which, I believe, should be re-
garded as an independent species. This Malaccan form, however, is
black, measures only 3.3-3.5 mm., and has the epinotum much less swol-
len, and apparently even less distinctly constricted off from the mesono-
tum than in stdJiogompha, and the postpetiole is sulcate dorsally. It
is, perhaps, admissible to regard stcthogompha as a subspecies of vacca.

*82a. Ceematogaster (Physocrema) stethogompha var. detri-

TiNODis, var. nov.

JVorkcr. Length 3.5 mm.

Differing from the typical form in having the head less distinctly concave
behind and in sculpture, the rugae of the head and thorax being finer, so that
the dense puncturation is more apparent and the vermiculate rugosity of the
thorax is laxer and finer, especially in the middorsal region. The petiole and
postpetiole are shining, the former elongate elliptical, the latter with a distinct
median furrow. Mandibles more yellowish than in the typical form.

A single worker from the Sarawak River, Kuching (H. W. Smith).

83. Crematogaster (Oxygyne) daisyi Forel.

Cremastogaster {Oxygyne) daisyi Forel, Ann. Soc. ent. Belgique, 1901, 45, p.
376, ^ 9 .

Type-locality: Sarawak (Haviland).

84. Crematogaster (Decacrema) decaaiera Forel.

Cremastogaster (Decacrema) decamera Forel, Ann. Soc. ent. Balgique, 1910, 54,
p. 18 nota, ^ 9 d'.

Type-locality: Sarawak f^Haviland).

Five workers and a dealated female from Kuching (John Hewitt),
"from Macaranga with slightly trifid leaves," and a winged female
from British North Borneo (E. B. Kershaw).

78 bulletin: museum of comparative zoology.

85. Crematogaster (Decacrema) borneensis Ern. Andre.

Cremastogaster borneensis Ern. Andre, Rev. ent., 1896, p. 263, y .
Type-locality: Borneo.

85a. Crematogaster (Decacrema) borneensis var. insulsa Forel.
Cremastogaster (Decacrema) borneensis var. insulsa Forel, Rev. Suisse zool.,
1911, 19, p. 33, y 9 d^.
Type-locality: Borneo (Hose).

85b. Crematogaster (Decacrema) borneensis var. harpyia Forel.

Cremastogaster {Decacrema) borneensis var. harpyia Forel, Rev. Suisse zool.,
1911, 19, p. 3.3, y 9.

Type-locality: Sarawak, Borneo (Haviland).

85c. Crematogaster (Decacrema) borneensis siibsp. symbia

Fprel.

Cremastogaster (Decacrema) borneensis subsp. symbia Forel, Rev. Suisse zool.,
1911, 19, p. 34, ^ 9 c^.

Type-locality: Sarawak, Borneo (Haviland).

85d. Crematogaster (Decacrema) borneensis subsp. novem

Forel.

Cremastogaster (Decacrema) borneensis subsp. novem Forel, Rev. Suisse zool.,
1911, 19, p. 35, S 9.

Type-locality: Sarawak, Borneo (Haviland).

I refer a single female from British North Borneo (E. B. Kershaw),
to this subspecies.

85e. Crematogaster (Decacrema) borneensis subsp. hosei Forel.

Cremastogaster (Decacrema) borneensis subsp. hosei Forel, Rev. Suisse zool.,
1911, 19, p. 35, y .

Type-locality: Sarawak, Borneo (Haviland).

wheeler: the ants of borneo. 79

85f. Grematogaster (Decacrema) borneensis subsp. capax Forel.

Cremastogasier {Decacrema) borneensis subsp. capax Forel, Rev. Suisse zool.,
1911, 19, p. 37, y, 9.

Type-locality: Sarawak, Borneo (Haviland).

86. Grematogaster (Decacrema) captiosa Forel.

Cremastogasier {Decacrema) captiosa Forel, Rev. Suisse zool., 1911, 19, p. 37, ^ .
Type-locality: Sarawak, Borneo (Haviland).

87. Grematogaster (Decacrema) angulosa Ern. Andre.

Cremastogasier angulosa Ern. Andre, Rev. ent., 1896, p. 264, S .
Type-locality: Borneo.

88. Grematogaster (Decacrema) biformis Ern. Andre.

Cremastogaster biformis Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 53, ^ .
Type-locality: Kapouas Basin, Borneo (Ghaper).

89. VOLLENHOVIA PUNCTATOSTRIATA Mayr.

Vollenhovia punctatostriata Mayr, Reise Novara. Zool., 2. Formicid., 1865, p. 21,
nota, 9 ; Mayi-, Jijdschr. ent., 1867, 10, p. 94, 9 ; Mayr, Ann. Mus. civ.
Genova, 1872, 2, p. 152, 9 ; Emery, ibid., 1887, ser. 2, 5, p. 453, 9 .

Type-locality: Java (Leyden Museum).
Sarawak fDoria and Beccari).

90. Vollenhovia rufiventris Forel.

Vollenhovia rufire^itris Forel, Ann. Soc. ent. Belgique, 1901, 45, p. 374, y .
Monomori'um riifirentre Emery, in Sarasin and Roux, Nova Caledonia. Zool.,
1914, 1, p. 407 nola, y .

Type-locality: Sarawak, Borneo (Haviland).
Female. Length 11.5 mm.

80 bulletin: museum of comparative zoology.

Head rectangular, slightly longer than broad, nearly as broad in front as
behind with feebly excised posterior border and a large shallow impression
on each side of the vertex. Eyes small, near the middle of the sides, ocelli
very small, the anterior distirctly larger than the posterior. Mandibles large,
with nearly straight external borders, the apical borders broad, deeply and
arcuately excised in the middle, with three large, subequal teeth at the apical
and three smaller, blunt teeth at the basal end of the excision. Clypeus very
short on the sides, rather flat in the middle, with a narrow elongation back-
ward between the frontal carinae, the anterior border straight and entire.
Frontal area obsolete; frontal carinae well-developed, nearly half as long as
the antennal scapes, slightly diverging behind, nearly as far apart as their
distance from the lateral border of the head. • Antennae short, 12-jointed,
the scapes strongly curved at the base, their tips reaching only to the posterior
orbits; the funiculi with a 3-jointed club; joints 2-5 broader than long, joint
6 as long as broad, the remaining joints longer than broad. Thorax rather
small, as broad as the head, the pronotum with submarginate sides and bluntly
rectangular humeri, the mesonotum flattened above, as long as broad, pro-
longed in the middle anteriorly; the epinotum short, in profile evenly rounded
and moderately convex, without distinct base and declivity, concave in the
middle behind between a pair of slight swellings representing the denticles of
other species. Petiole from above regularly rectangular, 1| times as long as
broad, as broad in front as behind, with sharp anterior and posterior angles;
in profile, with a node as high as the segment, its anterior surface concave, its
posterior convex, the summit blunt and rounded ; the lower surface anteriorly
produced as a large protuberance translucent in the middle, and tipped with
a small, blunt tooth. Postpetiole subrectangular, with convex sides and
dorsum, slightly broader than long and broader than the petiole, its ventral
surface with a large, thick, pointed, downwardly directed tooth at the anterior
end. Gaster elongate elliptical, narrowed anteriorly and posteriorly. Legs
with much thickened femora and clavate tibiae. Wings nearly 8 mm. long,
narrow, with a single cubital and a small, narrow discoidal cell; radial cell
open; pterostigma small but distinct.

Very shining; mandibles sparsely punctate, sharplj' longitudinally striate
along the external borders and at the tip. Clypeus uneven but not rugose.
Cheeks and anterior f- of head above, longitudinally and rather densely rugose,
with interspersed punctures; occipital region and sides of gula with coarse, pihg-
erous punctures. Pronotum coarsely and sparsely punctate, indistinctly striate
on the sides. Mesonotum finely, longitudinally rugose behind, anteriorly
smooth in the middle and coarsely and sparsely punctate on the sides. Scutel-
lum and sides of epinotum sharply, longitudinally rugose, the former smooth
and more or less coarsely punctate in the middle. Slope of epinotum, petiole,
postpetiole, gaster, scapes, and legs very smooth and shining, with sparse
piligerous punctures; lateral and ventral portions of petiole and postpetiole
densely and finely punctate.

Hairs slender, pointed, whitish, sparse, of very unequal length, suberect

wheeler: the ants of borneo. 81

or erect, shorter and more reclinate on the appendages, partly short and
appressed on the gaster.

Black; terminal antenna! joint, articulations of legs and tarsi, beyond the
basal joint, reddish brown; cheeks and tips of mandibles obscurely tinged with
red. Wings distinctly infuscated, more deeply along the anterior margin;
veins and pterostigma dark brown.

Described from a single specimen taken on ]Mt. Matang, West
Sarawak by G. E. Bryant and sent me by ]\Ir. Horace Donisthorpe.
There is also a worker from Kuching (John Hewitt) in my collection.

I have described the female in detail on account of its interest in
connection with Emery's contention (in Sarasin and Roilx, Nova
Caledonia. Zool., 1914, 1, p. 407 nota) that rufipentris is really a
Monomorium, mainly because the worker is smooth and shining and
has a slightl}^ pedunculate petiole. Although my specimen is not
accompanied by workers, Forel's recent discovery (Tijdschr. ent.,
1915, 58, p. 23) of all three phases of a new variety of this ant, var.
simahtrana from Simalur, a small island off the west coast of Sumatra,
shows that the Bornean female must belong to the tA-pical riifiirnfris
or one of its varieties. The worker simalurana varies from 3-7.9 mm.
in length, the female measures 12-12.5 mm. and the male only 4.4 mm.
These extraordinary differences in stature are comparable to those
previously noted by Forel (Philippine journ. sci., 1910, 5, p. 125) in
V. ohlonga subsp. dispar, the worker of which measures 3.2 mm., the
female 8 mm., the male 3.8 mm. Certainly the female of what I take
to be the t\'pical rufitentris described above and that of the var.
simalurana, which is merely somewhat larger and very slightly differ-
ent in other respects, must be regarded as belonging to Vollenhovia.
It is, moreover, closely related to I", sfriato punctata Mayr, known
only from the female (9-10 mm. long), and considered by Emery as
probably the female of V. ohlojiga subsp. laemthorax Emery (loc. cit.,
p. 406). In some species and subspecies of Vollenhovia, however,
the female is only slightly larger than the worker, e.g., in V. emeryi
Wheeler of Japan and in a Bornean subspecies of V. banksi Forel
described below.

91a. Vollenhovia oblonga Smith var. rufescens Emery.

Vollenhovia rufescens Emery, Bull. Soc. ent. France, 1894, p. 69, ^ .
Vollenhovia oblonga var. rufescens Emery, in Sarasin and Roux, Nova Cale-
donia. Zool., 1914, 1, p. 406, S .

Type-locality: Pulo Laut, Borneo.
Borneo (Coll. Emer}).

82 bulletin: museum of comparative zoology.

91b. Vollenhovia oblonga subsp. alluaudi Emery.

Vollenhovia alluaudi Emery, Bull. Soc. ent. France, 1894, p. 68, ^ .

Vollenhovia oblonga var. alluaudi Emery, Ann. Mus. civ. Geneva, 1897, ser.
2, 18, p. 560, ^ .

Vollenhovia oblonga subsp. alluaudi Emery, in Sarasin and Roux, Nova Cale-
donia. Zool., 1914, 1, p. 406; Forel, Trans. Linn. soc. London. Zool.j
1912, ser. 2, 15, p. 162, ^ 9 .

Type-locality: Seychelles.
Borneo (Coll. Emery).

91c. Vollenhovia oblonga subsp. laevithorax Emery.

Vollenhovia laevithorax Emery, Ann. Mus. civ. Genova, 1889, ser. 2, 7, p. 501, ^ •
Vollenhovia oblonga subsp. laevithorax Emery, in Sarasin and Roux, Nova
Caledonia. Zool., 1914, 1, p. 406.

Type-locality: Tenasserim (L. Fea).
Borneo (Coll. Emery).

*92. Vollenhovia hewitti, sp. nov.
Worker. Length 2.4-2.6 mm.

^&'

Head rectangular, about -J longer than broad, with parallel sides and nearly
straight posterior border. Eyes flattened, a little in front of the middle of the
sides. Mandibles rather broad, their apical and basal borders meeting at a
sharp right angle, the former with three large apical and a few minute and
indistinct basal teeth. Cl3T)eus rather convex in the middle, strongly bicari-
nate, its anterior border straight and entire. Frontal area small, semicircular,
frontal carinae short. Antennal scapes reaching nearly to the posterior third
of the head; funiculi with a 4-jointed club, joints 2-7 decidedly broader than
long. Thorax rather long, narrower than the head, the pro- and mesonotum
together longer than broad, with prominent humeri, somewhat flattened
dorsally and laterally; mesoepinotal constriction narrow but distinct. Epino-
tum from above a little longer than broad, distinctly narrower than the
mesonotum, laterally compressed above in profile, with the base feebly rounded,
and longer than the abrupt, concave declivity, which is slightly angulate but
not dentate on each side above. Petiole narrow, twice as long as broad, as
broad in front as behind, through the node nearly as high as long, the latter
blunt and rounded, with subequal, slightly concave anterior and feebly convex
posterior slope, the ventral surface anteriorly with a compressed, translucent

wheeler: the ants of borneo. S3

tooth. Postpetiole broader than the petiole, longer than broad, scarcely
broader behind than in front, elliptical, with convex lateral and dorsal surfaces.

Shining; mandibles with a few minute, scattered punctures; head with the
cheeks and anterior -g- above subopaque, longitudinally rugulose and sparsely
serially punctate; occiput and lower surface coarsely and sparsely punctate
and shining. Thorax above smooth, with small, scattered, piligerous punc-
tures; sides of thorax subopaque and densely punctate; extreme base of first
gastric segment sharply striate; remainder of body smooth and shining, with
indistinct, scattered, piligerous punctures.

Hairs pale, sparse, of unequal length, erect or reclinate; on the appendages
shorter and more appressed.

Dark castaneous brown; mandibles, tips of scapes, funiculi, trochanters,
knees, tibiae and tarsi red.

Described from four specimens taken at Kuching (John Hewitt).

This species resembles V. ohloncja siibsp. laevithorax Emery in
sculpture, but is much smaller, with narrower head, broader mandibles
and longer petiole and postpetiole, and is quite different from any of
the numerous other species of Vollenhovia recently described by
Emery, Forel, and Viehmeyer. The genus is difficult, so that the
limits of the subspecies and varieties are still to be established on the
basis of much more material than has been collected heretofore.

*93a. Vollenhovia banksi Forel subsp. kuchingensis, subsp. nov.
Worker,. Length 1.8-2 mm.

Differing from the typical form from the Philippines in its darker
color, the body, femora, and tibiae being dark brown. The meso-
epinotal impression is %asible, though very feeble ; the epinotum has a
minute denticle on each side, the antennal scapes are a little shorter
and the eyes a little further forward on the head.

Female (dealated). Length nearly 3 mm.

Resembling the worker, but the rugae on the head are coarser, more un-
dulating, with finely punctate interrugal spaces. The thorax above is coarsely
and rather densely punctate, the pronotum somewhat reticulatcly rugose, the
mesonotum smooth in the middle in front, longitudinally rugulose behind,
the mesopleurae rather smooth and shining in the middle. The epinotum is
coarsely reticulately rugose, the concavity of the declivity transversely rugose;
the petiole and postpetiole above coarsely foveolate. The tips of the anten-
nal scapes reach onlj^ a little beyond the median transverse diameter of the
head.

84 bulletin: museum of comparative zoology.

Described from twelve workers and a single female from Kuehing
(John Hewitt) taken "from an old fungus." I have compared the
workers "with a couple of cotypes of hanksi received from the Manila
Bureau of Science. Typc—M. C. Z. 8,949.

94. MoNOMORiuM FLORicoLA (Jerdon).

Atta floricola Jerdon, Madras Journ. lit. sci., 1851, 17, p. 107; Jerdon, Ann.

mag. nat. hist., 1854, ser. 2, 13, p. 49, y .
Monomnrium specuhre Ma>T, Sitzb. Akad. wiss. Wien, 1866, 53, p. 509, S .
Monomorium floricola Emery, in Dalla Torre, Cat. Hymenop., 1893, 7, p. 66.

Type-locality: Tellicherry, Southern India (Jerdon).

Sarawak (Doria and Beccari).

Several workers from Kuehing (John Hewitt).

95. MoNOMORiUM LATiNODE Mayr.

Monomorium latinode Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 152, ^ ;
Emery, ibid., 1887, ser. 2, 5, p. 459, y .

Type-locality: Sarawak (Doria and Beccari).

_96. MoNOMORiUM PHARAONis (Linne) .

For the long synonymy of this cosmopolitan ant see Dalla Torre, Cat. Hy-
menop., 1893, 7, p. 68.

Type-locality: Eg^ypt. *

Sarawak (Doria and Beccari).

Numerous workers and dealated females from Kuehing, Matang,
and Poi (John Hewitt).

*97. SoLENOPSis GEMiNATA (Fabricius) subsp. RUFA (Jerdon).

Atta rufa Jerdon, Madras Journ. lit. sci., 1851, 17, p. 106; Jerdon, Ann. mag.

nat. hist., 1854, ser. 2, 13, p. 48, Ql ^ 9 .
Soleiiopsis geminata Mayr, Verb. Zool. bot. gesellsch. Wien, 1870, 20, p. 996
(part); Mayr, ibid., 1886, 36, p. 460 (part); Rothney, Trans. Ent. soc.

London, 1889, p. 365.
S olenopsis geminata var. rufa Forel, Deutsch. ent. zeitsohr., 1909, p. 268.

wheeler: the ants of Borneo. 85

Soleywpsis geminata race rufa Forel, Biol. Centr. Amer. Hymenop., 1899-1900,
3, p. 80.

Type-locality: Malabar, Southern India (Jerdon).
Two workers and a male from Kuching (John Hewitt).

98. LoPHOMYRMEX BEDOTi Emery.

Lophomyrmex bedoti Emery, Rev. Suisse zool., 189.3, 1, p. 192, ^ 9 .

Type-locality: Deli, Sumatra (Bedot and Pictet).

Pulo Laut, Borneo (Coll. Emery).

A single worker from Kuching (John Hewitt).

99. Pheidologeton affinis (Jerdon).

Oecodoma affinis Jerdon, Madras Journ. lit. sci., 1851, 17, p. 110, Ql ^ ;

Jerdon, Ann. mag. nat. hist., 18.54, ser. 2, 13, p. 51, 21 y .
Pheidole affinis Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 174, ^ .
Atta bellicosa Smith, ibid., p. 164, ^ .
Solenopsis laboriosa Smith, Journ. Proc. Linn. soc. London. Zool., 1861, 6,

p. 48, y . .

Pheidologeton laboriosus Mayr, Verh. Zool. bot. geseilsch. Wien, 1862, 12,

p. 750, y .
Pheidologeton affinis Roger, Verzeich. formicid., 1863, p. 30.
Solenopsis calida Smith, Journ. Proc. Linn. soc. London. Zool., 1863, 7, p. 22,

y.
Pheidologeton bellicosum Mayr, Verh. Zool. bot. geseilsch. Wien, 1886, 36,
p. 360.

Type-locality: Malabar, Southern India (Jerdon).

Kapouas Basin, Borneo (Chaper).

A single worker media from Sarawak (Roland Thaxter).

100. Pheidologeton diversus (Jerdon).

Oecodoma diversus Jerdon, Madras Journ. lit. sci., 1851, 17, p. 109, S ; Jerdon,

Ann. mag. nat. hist., 1854, ser. 2, 13, p. 51, ^ .
Pheidole diversa Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 174, ^ .
Pheidole ocellifera Smith, ibid., p. 174, Ql .
Pheidole pabulator Smith, Journ. Proc. Linn. soc. London. Zool. 1861, 5,

p. 112, y .

86 bulletin: museum of comparative zoology.

Pheidologeton ocelliferus Mayr, Verh. Zool. bot. gesellsch. Wien, 1862, 12,

p. 750 Ql, pi. 19, fig. 13.
Pheidologeton megacephalus Roger, Verzeich. forraicid., 1863, p. 30 Ql .
Pheidologeton diversus Roger, ibid., p. 30.

Pheidologeton ocellifer Mayr, Reise Novara. Zool., 2. Formicid., 1865, p. 103.
Pheidologeton pabulator Mayr, Verh. Zool. bot. gesellsch. Wien., 1886, 36.

p. 362.
Pheidole megacephalotes Dalla Torre, Wien. ent. zeit., 1892, 11, p. 90.

Type-locality: Wynaad, Southern India (Jerdon).
Kapouas Basin, Borneo (Chaper).

101. Dilobocondyla borneensis Wheeler.

Dilobocondyla borneensis Wheeler, Proc. New Engl. zool. club, 1916, 6,
p. 12, g , fig. 2.

Type-locality: Bongo Mt., Sarawak, Borneo (John Hewitt).

102. Pristomyrmex trachylissus (Smith).

Myrmica (Monomorium) trachylissa Smith, Cat. Hymenop. Brit, mus., 1858,

6, p. 126, 9 .
Pristomyrmex trachylissa Mayr, Verh. Zool. bot. gesellsch. Wien, 1886, 36,

p. 359.
Pristomyrmex trachylissus Dalla Torre, Cat. Hymenop., 1893, 7, p. 62.

Type-locality: Borneo (A. R. Wallace).

103. Myrmecina undulata Emery.

Myrmecina undulata Emery, Ann. Mus. civ. Genova, 1900, ser. 2, 20, p. 678,

y 9.

Type-locality: Si Rambe, Sumatra (E. Modigliani).
Pulo Laut, Borneo (Coll. Emery).

*104. Acanthomyrmex dyak, sp. nov.

Soldier. Length 5 mm.

Head very large, convex above, overlapping the thorax nearly to the meso-
epinotal constriction, subrectangular, a little longer than broad, with rather

wheeler: the ants of borneo. 87

straight, subparallel sides, bilobed behind, owing to the deep, angular occipital
incision, which is continued into a deep occipital and frontal groove all the
way to the frontal area. Eyes small, moderately convex, elongate elliptical
at the anterior third of the head. Mandibles very large and convex, with
very broad, straight, edentate apical and very concave basal borders. Clypeus
depressed on the sides and behind, convex and roof -like, but not carinate in
the middle in front, where it projects over the proximal ends of the apical
mandibular borders as a short lobe, with straight margin and indistinctly
dentate corners. Frontal area rather distinct, triangular. Frontal carinae
strongly diverging behind, each bordering a deep narrow scrobe for the antennal
scape, running down obliquely to the side of the head a little behind and above
the eye. At this point it turns sharply at an angle to form a short scrobe for
the base of the funiculus, running forward just above the eye and merging
anteriorly into an interrugal space. Antennae slender, the scapes curved and
flattened but not dilated at the base. Funiculi with a 3-jointed club, shorter
than the remainder of the funiculus; -joints 2-8 as long as broad. Thorax
short, less than half as broad as the head; pro- and mesonotum rather de-
pressed dorsally where they are covered by the posterior portion of the head;
pronotum without spines. Mesoepinotal constriction not very pronounced.
Epinotum very short, its base much shorter than the vertical declivity; the
spines very long, slightly curved downward, thick at the base, strongly tapering
at the tips, diverging outward, backward and upward, much longer than their
distance apart at the base. Metasternal angles small, acute, directed upward.
Petiole from above nearly twice as long as broad, a little broader behind than
in front, with concave sides and narrow node, armed with two upwardly
directed acute spines, the space between which is semicircular; anterior
surface of the node long and concave, the posterior straight and abrupt.
Postpetiole a little broader than the petiole and a little broader than long,
very convex dorsally and rounded on the sides. Gaster only about ^ as large
as the head, subcircular, flattened, with straight anterior border. Femora
thickened in the middle, tibiae somewhat clavate.

Rather shining; mandibles smooth, with very fine, widely scattered punc-
tures. Clypeus smooth and shining. Head covered with umbilicate piliger-
ous foveolae, elongate on the gular surface, on the space between the frontal
carinae interspersed with strong, diverging longitudinal rugae; cheeks coarsely,
longitudinally rugose, space enclosed by the two limbs of the scrobes coarsely
reticulate-rugose; the scrobes smooth and shining, not transversely rugose.
Thorax coarsely and reticulately rugose; epinotal declivity between the spines
and the whole petiole smooth and shining; postpetiole coarsely and indis-
tinctly rugose above, subopaque. Gaster smooth and shining.

Hairs pale grayish, short, erect, blunt, but not thick, sparse, covering the
scapes and legs including the funiculi and tarsi as well as the body, more
abundant on the gaster; on the antennal scapes, most numerous on the anterior
surface; on the legs, most numerous on the extensor surfaces.

Deep ferruginous brown; mandibles, antennae, legs, epinotal spines, and
peduncle of petiole deep red; gaster and borders of mandibles blackish.

88 bulletin: museum of comparative zoology.

Worker. Length 3.5 mm.

Head much smaller than in the soldier, as broad as long, rather rectangular,
with straight posterior boi'der and feebly convex sides. Eyes small, but very
convex, hemispherical, distinctly behind the middle of the head. Mandibles
large, shaped somewhat as in the soldier, but with much less convex external
borders, with 2 or 3 apical and several minute, widely spaced basal teeth.
Clypeus moderately convex, with nearly straight anterior border. Frontal
area large, flat, triangular. Antennal scrobes short, limited to a groove for
the base of the scapes between the prominent rugae. Antennal scapes straight,
not flattened at the base, extending about ^ their length beyond the posterior
corners of the head. Funiculi also longer than in the soldier, with all the
joints longer than broad. Thorax shaped much as in the soldier, but the
pronotum with a pair of very long, straight, pointed spines, directed upward
and outward, distinctly longer than the similarly directed spines on the epino-
tum. The latter, however, are bent slightly backwards at a feeble angle just
beyond the middle. Shape of petiole, postpetiole, and gaster much as in the
soldier, but the spines on the petiole are less spreading and more erect, more
slender, and acute.

Sculpture of thorax like that of the soldier; mandibles and clypeus smooth,
the anterior border of the latter with short, coarse rugae, the head very coarsely
longitudinally rugose, the rugae connected by transverse rugules. Antennal
scrobes shining, feebly punctate, transversely rugulose anteriorly. Petiole
postpetiole and gaster smooth and shining.

Pilosity and color much like those of the soldier, except that the mandibles
are more yellowish and without dark borders.

Described from a single soldier and two workers taken by Mr. John
Hewdtt at Kuehing.

This species seems to be very close to A. for ox Emery, based on a
worker specimen from Perak. I am, however, unable to make Emery's
description of the sculpture of the head and thorax accord mth that of
the Bornean form. At any rate his words " f oveolis piligeris confertis
sculptum" do not seem to me to describe the conditions in my speci-
mens. Moreover, he gives the length of his specimen as about 4.5 mm.
and describes the two pairs of thoracic spines as equal (" thorax spinis
cjuatuor subrectis, aequalibus armatus"), and the postpetiole as
" sublaevis," whereas it is very smooth and shining in dyak. The ants
of the genus Acanthomyrmex seem to be very rare. No one has taken
A. notahilis since it was described by Smith, and Bingham in the
Fauna of British India merely translates Emery's description of A.
luciolae of Ceylon. Emery has recently based another species, A.
kochi, from New' Guinea (Nova Guinea, 9, Zool., 1911, 2, p. 252) on a

wheeler: the ants of borneo. 89

single worker specimen. It is very small (2.2 mm.), ferruginous yellow,
with the spines more curved and more nearly horizontal than in the
other species.

*105. ACANTHOMYRMEX DUSUN, Sp. nOV.

Soldier. Length 3.6 mm.

Head very large, extending back over the thorax to the mesoepinotal con-
■striction, subrectangular, a httle longer than broad, with straight, parallel
sides in front and rounded posterior corners, or lobes, separated by an angular
occipital incision continuous with a deep occipital and frontal groove, running
forward to the clypeus. In profile the dorsal surface of the head is convex
and rounded, but distinctly depressed in the middle just in front of the occi-
pital border. Eyes small, elongate elliptical, rather convex, at the anterior
fourth of the sides. Clypeus, mandibles, and antennal scrobes shaped much
as in dyak, but the frontal carinae which border them much more diverging
behind and the frontal area obsolete. Antennae also very similar, but joints
2 and 3 more transverse and joints 4-8 as long as broad. Thorax short, the
epinotal spines very long, stout at the base, tapering and pointed at the tips,
strongly curved backward and downward and somewhat outward. Petiole
fully twice as long as broad, its anteroposteriorly compressed node narrowed
above, without spines, but with its rather sharp, transverse margin angularly ex-
cised. Postpetiole regularly rectangular, broader than long, broader than the
petiole, rather flat above. Gaster about J as large as the head, excluding the
mandibles, nearly circular, with rather straight anterior border. Femora
thickened in the middle, tibiae clavate.

Shining; mandibles and clypeus smooth, the former sparsely and finely
punctate, the latter without rugosities along its anterior border. Head
covered rather uniformly with sparse, punctate foveolae, the triangular space
between the two limbs of each scrobe very coarsely reticulate-rugose, the space
between the frontal carinae finely, but not uniformly, longitudinally striate,
the scrobes transversely striolate. Thorax and postpetiole very coarsely
reticulate-rugose, epinotal declivity between the spines, the petiole and gaster
very smooth and shining.

Pilosity much as in dyak, but the hairs longer and more delicate, though
blunt, and somewhat more numerous on the head, forming an even row along
the anterior border of the antennal scapes and more numerous on the extensor
than on the flexor surfaces of the legs.

Brownish ferruginous; mandibles, antennae, and legs deep red; petiolar node,
postpetiole, and gaster black.

Described from a single specimen taken on Mt. Matang, West
Sarawak by Mr. G. E. Bryant and sent me by Mr. Horace Donis-
thorpe.

90 bulletin: museum of comparative zoology.

This species is certainly very different from dyak and luciolae, but
seems to be closely related to Smith's notabilis described from the
island of Batjan. Judging from Smith's figure and description,
however, dusun has a smaller and longer head and more curved epinotal
spines. My specimen seems to be too small to be the soldier of
Emery's jcrox.

106. Calyptomyrmex emeryi Forel.
Calyptomyrmex emeryi Forel, Mitth. Naturh. mus. Hamburg, 1901, 18, p. 51,

Type-locality: Sarawak, Borneo (Haviland).
Female. Length 4.5 mm.

Very similar to the worker. Thorax narrower than the head only slightly
longer than broad; mesonotum somewhat flattened above, subtriangular,
fully as broad as long; scutellum longitudinally impressed in the middle,
overhanging the extremely short, vertical epinotum. Both the mesonotum
and scutellum coarsely reticulate-rugose. Wings with well-developed discoidal
cell, large pterostigma and single cubital cell, the membranes uniformly
brownish hyaline, the veins and pterostigma darker brown.

A single specimen from Kuching (John Hewitt), which has been
compared wdth a worker cotype in my collection.

107. Meranoplus castaneus Smith.

Meranophis castaneus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 81, 9 , pi. 2, fig. 7; Forel, Zool. jahrb. Suppl., 1912, 15, p. 61.

Meranoplus cordatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 82, ^ , pi. 2, fig. 5; Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 155, ^ ;
Emery, ibid., 1887, ser. 2, 5, p. 470, S .

Type-locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari).

108. Meranoplus mucronatus Smith.

Meranoplus mucronatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857,
2, p. 82, S , pi. 2, fig. 6; Mayr, Tijdschr. ent., 1867, 10, p. 84, g .

Type-locality: Mt. Ophir, Malacca (A. R. Wallace).
Borneo.

wheeler: the ants of borneo. 91

*109. Tetramorium curtulum Emery.

Tciramorium curtuJum Emery, Ann. Mus. civ. Genova, 1894, ser. 2, 14, p. 22,

Typc-locolity: Pcalon, Biirmah (L. Fea).

A single worker from Mt. Matang, West Sarawak (G. E. Bryant),
received from Mr. Donisthorpe.

*110. Tetramorium simillimum (Smith).

Myrmica simiUima (Xylander) Smith, List. Brit. anim. Brit, mus., 1851, 6,

Acul. p. 118, ^ /
Tetrogmus caldarms Roger, Berl. ent. zeitschr., 1857, 1, p. 12, ^ 9 .
Myrmica (Leptothorax) simillima Smith, Cat. Brit. foss. Hymenop., 1858,

p.31, y.
Myrmica caldaria Meinert, Naturv. afh. Dansk. vid. selsk., 1860, ser. 5, 5,

p. 334, S 9 cT.
Tetramorium simillimum Mayr, Em-op. Formicid., 1861, p. 61, 9 .

Type-locality: Dorsetshire, England, in a hot-house (Dale).
Four workers from Sarawak (Roland Thaxter).

111. Tetramorium pacificum Mayr.

Tetramorium pacificum Mayr, Verh. Zool. bot. gesellsch. Wien, 1870, 20, p. 972,
976, y 9.

Type-locality: Tongatabu, Friendly Islands (Museum Godeffroy).
Kapouas Basin, Borneo (diaper); Tandjong (Fritz Suck).

112. Tetramorium scabrum Mayr.

Tetramorium scabrum Mayr, Verh. Zool. bot. gesellsch. Wien, 1878, 28, p.
672, y .

Type-locality: Borneo (Museum of Pest).

A single worker from Bongo Mt., Sarawak (Hewitt and Brooks).

113. Tetramorium guineense (Fabricius).

For the synonj'my of this common tropicopolitan ant see Dalla Torre, Cat.
Hymenop. 1893, 7, p. 133.

92 bulletin: museum of comparative zoology.

Type-locality: Guinea (Dr. Isert).

Sarawak (Doria and Beccari; Bedot and Pictet); Kapouas Basin
(Chaper).

Several workers and a female from Kuching (John Hewitt).

114. Triglyphothrix parvispina Emery.

Triglyphothrix 'parvispina Emery, Rev. Suisse zool., 1893, 1, p. 214, ^ .
Type-locality: Pulo Laut, Borneo.

*115. Triglyphothrix striatidens (Emery).

Tetramorium obesum st. striatidens Emery, Ann. Mus. civ. Genova, 1889, ser.

2, 7, p. 501, y .
Triglyphothrix obesus st. striatidens Emery, E.xplor. sci. Tunisie. Fourmis,

1891, p. 4.
Triglyphothrix striatidens Dalla Torre, Cat. H3^menop., 1893, 7, p. 136.

Type-locality: Bhamo, Burmah (L. Fea).

A worker and female from Kuching f John Hewitt) and a worker
from Sarawak (Roland Thaxter).

116. Cataulacus brookei Forel.

Cataulacus brookei Forel, Ann. Soc. ent. Belgique, 1901, 45, p. 378, ^ 9 c?.
Type-locality: Sarawak, Borneo (Haviland).

117. Cataulacus granulatus (Latreille).

Formica gramdata Latreille, Hist. nat. fourmis, 1802, p. 275, ^ , pi. 12, fig. 75.
Cryptocerus granulatus Lepeletier, Hist. nat. ins. hymenop. 1836, 1, p. 171.
Cataulacus granulatus Smith, Trans. Ent. soc. London, 1854, ser. 2, 2, p. 226.
Cataulacus reticulatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 81, y , pi. 2, fig. 8.

Type-locality: "Grandes-Indes."
Sarawak, Borneo (A. R. Wallace).

wheeler: the ants of Borneo. 93

117a. Cataulacus granulatus subsp. hispidus Smith.

Caiaulacus hispidus Smith, Trans. Ent. soc. London, 1876, p. 611, ^ , pi. 11,

fig. 11.
Cataulacus granulatus subsp. hispidus Forel, Rev. Suisse zool., 1911, 19, p. 24,

Type-locality: Singapore.

Sarawak, Borneo (Haviland).

A single worker from Kuching (John Hewitt),

118. Cataulacus hispidulus Smith.

Cataulacus hispidulus Smith, Jouj'n. Proc. Linn. soc. London. Zool., 1864, S,

p. 76, y , pi. 4, fig. 7; Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 155 ;

Emery, ibid., 1887, ser. 2, 5, p. 470, ^ ; Em. Andre, Mem. Soc. zool.

France, 1892, 5, p. 55, ^ .
Cataulacus granulatus var. hispidtdus Dalla Torre, Cat. Hymenop., 1893, 7,

p. 138.

Type-locality: Sumatra (A. R. Wallace).
Sarawak (Doria and Beeeari), Kapouas Basin (Chaper).
Four workers from Kuching (John Hewitt) and one from Bongo Mt.
(Hewitt and Brooks).

119. Cataulacus horridus Smith.

Cataulacus horridus Smith, Journ. Proc. Linn. Soc. London. Zool., 1857, 2,
p. 81, y , pi. 2, fig. 3; Emery, Rev. Suisse zool., 1893, 1, p. 216.

Cataulacus insularis Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 80, &, pi. 2, fig. 4.

Type-locality: Sarawak, Borneo (A. R. Wallace).

Three workers and six males from Kuching (John Hewitt). The
males agree very closely with Smith's description of his C. insularis,
except in having the gaster entirely black, so that I regard insularis as
a synonjon of horridus. He describes the gaster of his specimen as
reddish at the base, a condition probably due to immaturity.

120. Cataulacus praetextus Smith.

Cataidacus praetextus Smith, Trans. Ent. soc. London, 1867, ser. 3, 5, p. 528, ^ ,
pi. 26, fig. 5.

94 bulletin: museum of comparative zoology.

Type-locality: Borneo.

I refer a female specimen from Kuching (John Hewitt) to this
species.

*121. Cataulacus latissimus Emery.

Cataulacus latissimus Emery, Rev. Suisse zooL, 1893, 1, p. 21,5, S , pi. 8, fig. 10.

Type-locality: Perak (Bedot and Pictet).
Two workers from Kuching (John Hewitt).

*122. Strumigenys indagatrix, sp. nov.

Worker. Length 1.7 mm.

Slender; head flattened, If times as long as broad, with semicircularly
excised posterior border, narrowed at the eyes, the portion anterior to the
antennal insertions as long as broad; eyes moderately large and convex.
Mandibles about half as long as the head, straight, slender, slightly narrowed
at their insertions, with straight, parallel borders, a pair of long, acute, subequal
apical teeth and a spiniform preapical tooth of the same shape but somewhat
shorter. Clypeus as long as broad, flat, with entire anterior border. An-
tennae slender; the scapes reaching nearly to the posterior fifth of the head;
last funicular joint longer than the remainder of the funiculus; joints 2 and 3
subequal, small, not longer than broad, fourth joint a little longer than the
first, about twice as long as the second and third together. Thorax narrow,
the profile outline of the pronotum feebly convex, rising to the mesonotum,
■which is the highest part of the thorax, and descending gradually to the long
epinotum, the base of which is nearly straight in profile and fully If times as
long as the declivity. The spines are rather long and slender, but indistinct,
as each is embedded in a spongiform plate continued down the side of the
declivity. Petiole and postpetiole with large spongiform masses on the sides
and ventrally, so that only the dorsal surfaces are exposed. Gaster but little
broader than its straight, transverse, anterior border. Legs slender.

Mandibles, gaster, and dorsal surface of postpetiole smooth and shining;
head, thorax, petiole, legs, and antennae opaque, densely and regularly punctate,
the head a little more coarsely than the thoracic dorsum, the appendages very
finely. Extreme base of gaster longitudinally rugose.

Hairs yellowish, sparse, rather long, especially at the tip of the gaster, erect
or suberect, shorter and blunter on the head, on the cljqieus and anterior
border of the scape curved but pointed, nonclavate. Legs with very short,
subappressed hairs.

Yellowish ferruginous; mandibles and legs paler.

wheeler: the ants of borneo. 95

Described from two specimens taken by Mr. John Hewitt at Kuch-
ing.

In the shape of the head, mandibles, and antennae this species
approaches S. viocsaryi Emery of Papuasia, but the mandibles are
decidedly shorter and the pilosity and arrangement of the spongiform
appendages of the petiole are very different. The Bornean species
is also related to four Javanese species recently described by Forel
{S. kraepelini, koningsbergi, signcae, and ehhae), but is smaller than any
of them. It can be readily distinguished from kraepelini and ko7iings-
bergi by the well-developed, spiniform, preapical tooth of the mandi-
bles. S. signeae possesses a translucent inner border to the mandibles
and an emarginate clypeus and in cbbae the preapical tooth is only
slightly longer than broad, the antennal scapes are shorter, the first
funicular joint not longer than broad, the hairs are club-shaped and
the color is deep ferruginous.

*123. Strumigenys bryanti, sp. nov.

Female. Length, 4.8 mm.

Slender; head fully If times as long as broad, its posterior border deeply
and semicircularly excised, the sides gradually converging to the rather large,
convex eyes, the preocular portion 1 § times as long as broad, with subparallel
sides; vertex very feebly convex, the posterior lobes depressed and flattened.
Mandibles fully | as long as the head, straight, flattened, with parallel borders,
very slightly narrowed at their insertions, sinuately excised at the inner border
just back of the two strong, subequal, spiniform, apical teeth. Preapical
tooth absent. Clypeus flattened, as long as broad, its anterior border entire.
Antennae very slender; scapes reaching to the posterior fourth of the head;
last funicular joint much shorter than the remainder of the funiculus and some-
what longer than the preceding joint, joints 2 and 3 subequal, as long as broad,
together a little more than half as long as the first joint. Thorax distinctly
narrower than the head, a little more than twice as long as broad, very high
and convex in the region of the mesonotum and scutellum, the epinotum low
with short horizontal base and long vertical declivity, meeting it at a right
angle; spines reduced to rectangular flat teeth, each continued as a trans-
lucent, spongiform plate down the side of the declivity. Petiole three times
as long as broad, broader behind than in front, with straight sides, the node
low and rounded, in profile gradually passing into the peduncle, its posterior
border with spongiform appendages and its ventral surface behind with a
median spongiform lamella. Postpetiole subcircular, as long as broad,
broader than the petiolar node, only its dorsal surface exposed, the remainder
embedded in spongiform material, with a long, flat plate of the same substance

96 bulletin: museum of comparative zoology.

depending from its midventral line. Gaster small, narrow and straight in
front, its antero ventral surface with a broad mass of spongiform material.
Legs long and slender. Wings with small pterostigma and almost no veins.

Opaque; mandibles and gaster shining; the gaster longitudinally rugose
and reticulate at the base. Remainder of body densely and finely punctate,
rugulose, the posterior lobes of the head regularly and more coarsely, the
appendages more finely punctate.

Hairs pale yellow; very long, sparse, erect, slender, and pointed, quite as
long and conspicuous on the legs as on the body; antennal scapes with short
curved, subappressed, pointed hairs, not arranged in a regular row. Head,
clypeus, and thorax also with numerous short, appressed, simple hairs, repre-
senting a long, dilute pubescence.

Ferruginous brown; mandibles, antennae, legs, and base and tip of gaster
paler and more yellowish. Wings yellowish hyaline with dark brown ptero-
stigma.

Described from a single specimen taken by Mr. G. E. Bryant on
Mt. Matang, West Sarawak and sent me by Mr. Horace Donisthorpe.

In size, in the general shape of the body and its extraordinary
pilosity, this species is most closely related to S. doriae Emery, de-
scribed from a worker specimen taken in Amboina. This form, how-
ever, has more slender and cylindrical mandibles, so that the Bornean
specimen can hardly be the cospecific female.

*124. Rhopalothrix borneensis, sp. nov.

Female (dealated). Length, 3.5 mm.

Resembling Rh. procera Emery, but smaller and the emarginations of the
sides of the head at the eyes and antennal insertions deeper and more sinuous
and the border just behind the ej'es forming a rounded, rather flattened lobe.
The clypeus is distinctly longer, its median portion being prolonged further
posteriorly. Thorax decidedly narrower than the head, the pronotum later-
ally compressed, deeply and angularly excised behind. Mesonotum flattened
above, grooved in the middle, but not carinate anteriorly. Scutellum and
epinotum very small, the former gibbous, overhanging the latter, which has a
pair of acute, flattened teeth as long as broad at their bases. Petiolar node
subrectangular when seen from above, as long as broad; feebly longitudinally
impressed in the middle. Postpetiole 2h times as broad as the petiole, much
broader than long, with broadly concave anterior and broadly convex posterior
border, the dorsal surface with a feeble median sulcus in front and a large
round impression behind. Gaster with a faint median sulcus.

Opaque; densely and finely punctate-rugulose; gaster and postpetiole
evenly punctate; antennal scrobes and mesopleurae shining.

wheeler: the ants of borneo. 97

Head, legs, and dorsal surface of body covered with dirty white, flattened,
appressed, scale-like hairs, fine and numerous on the clypeus and mandibles,
long and conspicuous on the external borders of the antennal scapes, tibiae,
and tarsi, absent on the pleurae and lateral surfaces of the coxae.

Ferruginous brown; appendages scarcely paler.

Described from a single specimen taken by Mr. John Hewitt at
Kuching in 1908 "in an old fungus."

Additional material may show that this is merely a small subspecies
of Rh. yroccra Emery of New Guinea. The female of this form
measures 5 mm. and has the petiolar node broader than long. Rh.
borneensis seems also to be more opaque than procera, but Emery's
description of the female is very brief.

DOLICHODERINAE.

125. DoLiCHODERUs (Hypoclinea) bituberculatus (Mayr).

Hypoclinea bituberculata Mayr, Verb. Zool. bot. gesellsch. Wien., 1862, 12,

p. 705, ^ .
Dolichoderus bituberculatus Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 4,

p. 254, y 9 .
Hypoclinea sellaris Roger, Berl. ent. zeitschr., 1863, 7, p. 168, 213, ^ .
Dolichoderus {Hypoclinea) bituberculatus Emery, Gen. Ins. Dolichod., 1912, p. 10.

Type-locality: Luzon Island, Philippines.

Sarawak (Doria and Beccari; Bedot and Pictet); Kapouas Basin
(Chaper); Bandjermassin (Fritz Suck).

Numerous workers from British North Borneo (E. B. Kershaw) and
from Serambu Mt., Sarawak (H. W. Smith). A dealated female and
five workers from Kuching (John Hewitt) are of a more reddish color.

12oa. Dolichoderus (Hypoclinea) bituberculatus var. borne-

ONENSis (Roger).

Hypoclinea sellaris var. borneonensis Roger, Berl. ent. zeitschr., 1863, 7, p.
214, S .

Type-locality: Borneo. •

126. Dolichoderus (Hypoclinea) coniger Mayr.

Hypoclinea conigera Mayr, Verb. Zool. bot. gesellsch. Wien, 1870, 20, p. 956, g .
Dolichoderus {Hypoclinea) coniger Emery, Gen. Ins. Dolichod., 1912, p. 13.

98 bulletin: museum of comparative zoology.

Type-locality: Sarawak, Borneo (Doria and Beccarl).

A single worker from Mt. Matang, West Sarawak (John Hewitt).

127. Dolichoderus (Hypoclinea) cuspidatus (Smith).

Polyrhachis cuspidatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857,

2, p. 63, S .
Hypoclinea cuspidMa Mayr, Verh. Zool. bot. gesellsch. Wien, 1870, 20,

p. 955, ^ .
Dolichoderus cuspidatus Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 4,

p. 2.56, ^ .
Dolichoderus {Hypoclinea) cuspidatus Emery, Gen. Ins. Dolichod., 1912, p. 13.

Type locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari).

Several workers from Mt. Matang and the Rambungan River,
Sarawak (H. W. Smith).

128. Dolichoderus (Hypoclinea) patens (Mayr).

Hypoclinea patens Mayr, Verh. Zool. bot. gesellsch. Wien, 1870, 20, p. 957, S .
Dolichoderus patens Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 4, p. 254.
Dolichoderus {Hypoclinea) patens Emery, Gen. Ins. Dolichod., 1912, p. 14.

Type-locality: Sarawak, Borneo (Doria and Beccari).

129. Dolichoderus (Hypoclinea) semirugosus (Mayr).

Hypoclinea semirugosa Mayr, Verh. Zool. bot. gesellsch. Wien, 1870, 20,

p. 956, y .
Dolichoderus semirugosus Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 4,

p. 254, ^ .
Dolichoderus {Hypoclinea) semirugosus Emery, Gen. Ins. Dolichod., 1912,

p. 14.

Type-locality: Sarawak, Borneo (Doria and Beccari).

130. Dolichoderus (Hypoclinea) sulcaticeps (Mayr).

Hypoclinea sulcaticeps Mayr, Verh. Zool. bot. gesellsch. Wien, 1870, 20,

p. 957, y .
Dolichoderus sulcaticeps Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 4,

p. 254, ^ .
Dolichoderus {Hypoclinea) sulcaticeps Emery, Gen. Ins. Dolichod., 1912, p. 14.

wheeler: the ants of borneo. 99

Type-locality: Sarawak, Borneo (Doria and Beccari).
Several workers from Sadong, Sarawak (H. W. Smith).

*131. DoLiCHODERUs (Hypoclinea) taprobanae (Smith).

Formica taprobane Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 13, 9 .
Dolichoderus taprobanae MajT, Verh. Zool. hot. gesellsch. Wien, 1886, 36, p. 353.
Hypocli7iea gracilis Motschulsky, Bull. Soc. nat. Moscou, 1863, 36, p. 14, ^ .
Dolichoderus (Hypoclinea) taprobanae Emery, Gen. Ins. Dolichod., 1912, p. 14.

Type-locality: Ceylon.

Two workers from Kuching (John Hewitt).

131a. Dolichoderus (Hypoclinea) taprobanae (Smith) var.

borneensis Forel.

Dolichoderus taprobanae var. borneensis Forel, Rev. Suisse zool., 1911, 19,
p. 46, y .

Dolichoderus (Hypoclinea) taprobanae var. borneensis Emery, Gen. Ins. Doli-
chod., 1912, p. 14.

Type-locality: Hayvep, Borneo (Winkler).

132. Iridomyrmex anceps (Roger).

Formica anceps Roger, Berl. ent. zeitschr., 1863, 7, p. 164, S .
Iridomyrmex excisus Mayr, Tijdschr. ent., 1867, 10, p. 77, § 9 , pi. 2, fig. 8.
Hypoclinea (Iridomyrmex) excisa Mayr, Verh. Zool. hot. gesellsch. Wien, 1870,

20, p. 959, y .
Irvlomyrmex anceps Emery in Dalla Torre, Cat. Hymenop., 1893, 7, p. 169;

Emery, Gen. Ins. Dolichod., 1912, p. 23.

Type-locality: Malacca.

Sarawak (Doria and Beccari).

Several workers and three females from Kuching (John Hewitt).

133a. Iridomyrmex cordatus (Smith) subsp. protensus Forel.

Iridomyrmex cordatus subsp. protensus Forel, Rev. Suisse zool., 1911, 19,
p. 47, y 9 d'; Emery, Gen. Ins. Dolichod., 1912, p. 24.

Type-locality: Sarawak, Borneo (Haviland).

100 bulletin: museum of comparative zoology.

*134a. Iridomyrmex myrmecodiae Emery.

Iridomyrmez cordatus var. myrmecodiae Emery, Ann. Mus. civ. Genova, 1887,

ser. 2, 4, p. 249, U .
Iridomyrmex myrmecodiae Emery, ibid., 1900, ser. 2, 20, p. 694; Emery, Gen.

Ins. Dolichod., 1912, p. 25.

Type-locality: Java.

Numerous workers from British North Borneo (E. B. Kershaw)
and a dealated female and two workers from Kuching (John Hewitt),
found nesting in the rootstoeks of a fern {Polypodium sinuosus
sinuosus).

135. Tapinoma flavidum Em. Andre.

Tapinoma flavidum Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 51, ^ ;
Emery, Gen. Ins. Dolichod., 1912, p. 41.

Type-locality: Kapouas Basin, Borneo (Chaper).

136. Tapinoma melanocephalum (Fabrieius).

Formica melanoccphala Fabrieius, Ent. syst., 1793, 2, p. .353, 9 .

Lasius melanocephalus Fabrieius, Sj^st. Piez., 1804, p. 417.

Myrmica melanoccphala Lepeletier, Hist. nat. ins. hymenop., 1836, 1, p. 185.

Formica nana Jerdon, Madras Journ. lit. sci., 1851, 17, p. 125, ^ .

Micromyrma melanocephnla Roger, Berl. ent. zeitschr., 1862, 6, p. 258, ^ 9 .

Myrmica pellucida Smith, Journ. Froc. Linn. soc. London. Zool., 1857, 2,

■ p. 71, y .
Formica familiaris Smith, ibid., Suppl., 1860, 4, p. 96 {nee. ibid., p. 68, 9).
Tapinoma melanocephalum Mayr, Verb. Zool. bot. gesellsch. Wien, 1862, 12,
p. 651; Emery, Gen. Ins. Dolichod., 1912, p. 41.

Type-locality: Cayenne (Coll. Bosc.).
Kapouas Basin (Chaper).

*137. Tapinoma indicum Forel.

Tapinoma melanocephalum var. Mayr, Termesz. fiizetek., 1897, 20, p. 432, y .
Tapinoma indicum Forel, Journ. Bombay nat. hist, soc, 1895, 9, p. 472, y ;

Bingham, Fauna Brit. India. Hymenop., 1903, 2, p. 304, ^ ; Emery,

Gen. Ins. DoHchod., 1912, p. 41.

wheeler: the ants of Borneo. 101

Ta'pinoma melanocephahim subsp. indicum Forel, Notes Leyden mus., 1911, 33,
p. 206.

Type-locality: Poona, India (Wroughton).
Several workers from Sarawak (Roland Thaxter).

138. Technomyrmex albipes (Smith).

Formica (Tapinoma) albipes Smith, Journ. Proc. Linn. soc. London. Zool.,

1861, 6, p. 38, S .
Tapinoma albipes Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 147; Emery,

ibid., 1887, ser. 2, 4, p. 249.
Technomyrmex albipes Emery, Zeitschr. wiss. zool., 1888,. 46, p. 392; Emery,

Gen. Ins. Dolichod., 1912, p. 43.

Type-localiiy: Tondano, Celebes (A. R. Wallace).
Sarawak (Doria and Beccari); Kapouas Basin (Cliaper).
Several workers from Sarawak (H. W. Smith) and one from Kuching
(John Hewitt).

139. Technomyrmex strenuus Mayr.

Technomyrmex strenua Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 147, S 9 .
Technomyrviex strenuus Emery, ibid., 1887, ser. -2, 4, p. 248, y ; Emery, Gen.
Ins. Dolichod., 1912, p. 44, g 9 , pi. 1, fig. 20, 20b.

Type-locality: Sarawak, Borneo (Doria and Beccari).

Sarawak (Haviland).

Four workers from Kuching (John Hewitt).

Camponotinae.

140. Myrmoteras donisthorpei Wheeler.

Myrmoteras donisthorpei Wheeler, Proc. New Eng. zool. club, 1916, 6, p. 14, 9 ,
fig. 3.

Type-locality: Mt. Matang, West Sarawak (G. E. Bryant).

141. Plagiolepis (Anoplolepis) longipes (Jerdon),

Formica longipes Jerdon, Madras Journ. lit. sci., 1851, 17, p. 122, ^ .
Formica gracilipes Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 55, y .

102 bulletin: museum of comparative zoology.

Formica trifasciata Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 27, 9 .
Prenolepis gracilipes Mayr, Verb. Zool. bot. gesellsch., Wien, 1862, 12, p. 698.
Plagiolepis gracilipes Mayr, Tijdschr. ent., 1867, 10, p. 73, ^ 9 .
Plagiolepis longipes Emery, Ann. Mus. civ. Geneva, 1887, ser. 2, 4, p. 247.
Plagiolepis (Anoplolepis) longipes Forel, Tijdschr. ent., 1915, 58, p. 34.

Type-locality: India (Jerdon).

Sarawak (Doria and Beccari; Bedot and Pictet); Kapouas Basin
(Chaper).

Numerous workers from Sandakan, Sadong and Mt. Matang,
Sarawak (H. W. Smith), British North Borneo (E. B. Kershaw) and
Kuching (John Hewitt).

142. Aphomomyrmex andrei (Emery).

Dimorphomyrmex andrei Emery, Ann. Sac. ent. France, 1894, p. 73, 9 .
Aphomomijrmex andrei Emery, Ann. Soc. ent. Belgique, 1899, 43, p. 894, 9 I
Wheeler, Psyche, 1910, 17, p. 132, 9 .

Type-locality: Pulo Laut, Borneo.

143. Aphomomyrmex hewitti Wheeler.
Aphomomyrmex hewitti Wheeler, Psyche, 1910, 17, p. 132, ^ 9, fig. 1.

Type-locality: Bidi, Borneo (John Hewitt).

144. Gesomyrmex chaperi Ern. Andre.

Gesomyrmex chaperi Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 47, ^ .
fig. 1-3.

Type-locality: Kapouas Basin, Borneo (Chaper).

145. Dimorphomyrmex janeti Enr. Andre.

Dimorphomyrmex janeti Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 51, ^ ,
fig. 4, 5; Wheeler, Psyche, 1910, 17, p. 132, g .

Type-locality: Kapouas Basin, Borneo (Chaper).

wheeler: the ants of borneo. 103

*146. Prenolepis jerdoni Emery.

Prenolepis jerdoni Emery, Rev. Suisse zooL, 1893, 1, p. 223, ^ , pi. 8, fig. 20.

Type-locality: Perak.

Three workers from Kuching (John Hewitt).

*147. Prenolepis (Nylanderia) longicornis (Latreille).

Formica longicornis L&treiWe, Hist. nat. fourmis, 1802, p. 113, S .
Formica vagans Jerdon, Madras Journ. lit. sci., 1851, 17, p. 124, ^ 9 .
Formica {Tapinoma) gracilescens Nylander, Ann. sci. nat. ZooL, 1856, ser. 4,

5, p. 73, ^ , pi. 3, fig. 20.
Formica gracilescens Nylander, Ann. Soc. ent. France, 1856, ser. 3, 4, Bull.,

p. XXVIII, S .
Tapinoma gracilescens Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 56.
Prenolepis gracilescens Mayr, Verb. ZooL bot. gesellsch. Wien, 1862, 12, p. 698.
Paratrechia currens Motschulsky, Bull. Soc. nat. Moscou, 1863, 36, p. 14.
Prenolepis longicornis Roger, Verzeichn. formicid., 1863, p. 10.
Prenolepis (Nylanderia) longicornis Santschi, Voy. Allmand et Jeamel Afr.

Orient. Formicid., 1914, p. 127, ^ .

Type-locality: Senegal (Bosc.).

Seven workers from Sarawak (Roland Thaxter).

148. Prenolepis (Nylanderia) obscura Mayr.

Prenolepis obscura Mayr, Verh. ZooL bot. gesellsch. Wien, 1862, 12, p. 698,

g 9.

Type-locality: Sydney, New South Wales (Novara Expedition).
Kapouas Basin, Borneo (Chaper).

*149. Prenolepis (Nylanderia) kraepelini Forel.

Prenolepis kraepelini Forel, Mitth. Naturh. mus. Hamburg, 1905, 22, p. 24,

y 9 cf.

Type-locality: Buitenzorg, Java (Kraepelin).

Two workers from Kuching (John Hewtt), "nesting in fungus,"
and one from Sarawak (Roland Thaxter). Another worker from Mt.
Matang (G. E. Bryant) appears to belong to the same species but is
considerably larger (2.6 mm.).

104 bulletin: museum of comparative zoology.

150a. Prenolepis (Nylanderia) butteli Forel subsp. bryantI

Forel.

Prenolepis {Nylanderia) hutteli Forel subsp. bryanti Forel, Rev. Suisse zool.
1916, 24, p. 439, ^ .

Type-locality: Mt. Matang, West Sarawak (G. E. Bryant).
Worker. Length 1.5 mm.

Head subrectangular, as broad as long, nearly as broad in front as behind
with rounded sides and posterior corners and feebly convex posterior border.
Eyes moderatel}^ large, flat, just behind the middle of the sides. Mandibles
narrow, with straight external and short, minutely 5-toothed apical borders.
Clypeus very convex, but not carinate, its anterior border entire and rounded.
Frontal carinae short, as far apart as their distance from the sides of the head.
Antennae slender, scapes extending about ^ their length beyond the occipital
border of the head; second funicular joint small, broader than long, joints
3-6 a little longer than broad. Thorax short, with distinct promesonotal
and mesoepinotal sutures, very feeble and short mesoepinotal constriction, the
pro- and mesonotum rather flat ; the epinotum with very short and transverse,
feebly convex base, passing rapidly into the long sloping declivity. Petiole
with very short, low node, a mere anteriorly directed projection at the anterior
end of the segment, much as in some species of Tapinoma. Gaster with the
first segment concave in front and lying over the petiole.

Head and thorax subopaque; mandibles, legs, antennae, and gaster more
shining; mandibles sparsely and finely punctate; the head, thorax, and gaster
distinctly shagreened, the gaster transversely.

Hairs dark brown, coarse, sparse, erect, blunt, except on the anterior portion
of the head; body without, antennae and legs with extremely short, appressed,
whitish pubescence.

Castaneous brown; head, thorax, and gaster with pronounced metallic
reflections, the head and pronotum more blue, the gaster and remainder of the
thorax more violet. Mandibles, antennae, tarsi, middle and hind tibiaef
knees, and ends of tibiae of fore legs, and tips of hind femora whitish yellow.

A single specimen from Sarawak, received from Prof. Roland
Thaxter.

Forel's description is very brief and is hidden away in the midst
of a paper devoted to Congolese ants!

151. Pseudolasius mayri Emery.

Lasius familiaris Mayr, Reiss Novara. Zool., 2. Formicid., 186.5, p. 55, 9
{nee Smith).

wheeler: the ants of Borneo. 105

Pseudolasiiis familiaris Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 4, p. 245,

nota, ^ {nee Smith nee 9 ).
Pseudolasiiis mayri Emery, Ann. Soc. ent. Belgique, 1911, 55, p. 215, § 9 cf.

Type-locality: Buitenzorg, Java.
Sarawak (Coll. Emery).

151a. Pseudolasius mayri var. sarawakanus Forel.

Pseudolasius mayri var. sarawakana Forel, Zool. jahrb. Syst., 1913, 36, p. 107
nota, 9 d'.

Type-locality: Sarawak, Borneo (Ha\aland).

152. Oecophylla smaragdina (Fabricius).

Formica smaragdina Fabricius, Syst. ent., 1775, p. 828, 9 .

Formica maera Guerin, Duperry Voy. Coquille. Zool., 1830, 2, p. 202, ^ ,

pi. 8, fig. 1.
Formica zonata Guerin, ibid., p. 205, 9 .
Oecophylla smaragdina Smith, Journ. Proc. Linn. soc. London. Zool. Suppl.,

1860, 4, p. 102, ^ .

Type-locality: India (Edier).

Sarawak (Doria and Beccari; Bedot and Pictet; A. R. Wallace).

Numerous workers from British North Borneo (E. B. Kershaw);
and a deiilated female and workers from Kuching and Sadong (H. W.
Smith).

153. Calomyrmex laevissimus (Smith).

Formica laevissima Smith, Journ. Proc. Linn. soc. London. Zool., 1859, 3,

p. 138, y .

Camponotus laevissimus Mayr, Tijdschr. ent., 1867, 10, p. 39, ^ , pi. 2, fig. 1.
Colobopsis levissima Roger, Verzeich.-- formicid., 1863, p. 10.
Calomyrmex laevissimus Emery, Mem. R. accad. sci. Bologna, 1896, 1895-1896,
ser. 5, 5, p. 776.

Type-locality: Aru (A. R. Wallace).
Borneo {teste Roger).

154. Camponotus (Dinomyrmex) gigas (Latreille).

Formica gigas Latreille, Hist. nat. fourmis, 1802, p. 105, 9 , pi. 2, fig. 6.
Camponotus gigas Mayr, Verh. Zool. bot. gesellsch. Wien, 1862, 12, p. 669, S .
Camponotus {Dinomyrmex) gigas Forel, Rev. Suisse zool., 1914, 22, p. 268.

106 bulletin: mi'seum of comparative zoology.

Type-locality: "Grandes-Indes" (Riche).

Several workers from British North Borneo (E. B. Kershaw) and
from the Sihmpopon River (Hugh M. Smith).

154a. Camponotus (Dinomyrmex) gigas subsp. borneensis

Emery.

Camponotus gigas subsp. borneensis Emery, Ann. Mus. civ. Geneva, 1887,
ser. 2, 4, p. 214, § 9 a"'

Type-locality: Sarawak, Borneo (Doria and Beccari).

Kapouas Basin (Chaper); Poeloe Island, E. Borneo (P. Jachan).

Numerous workers from Kuching, Sadong, and the Rambungan
River, Sarawak (H. W. Smith) and a single worker from Kuching
(John Hewitt).

155. Camponotus (Dinomyrmex) angusticollis (Jerdon).

Formica angusticollis Jerdon, Madras Journ. lit. sci., 1851, 17, p. 120, Qi ^ 9 .

Formica ardens Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 17, 9 .

Formica impetuasa Smith, ibid., p. 18, U .

Formica callida Smith, ibid., p. 18, S .

Camponotus prismaticus Mayr, Verb. Zool. bot. gesellsch. Wien, 1862, 12,

p. 669, U .
Camponotus angusticollis Roger, Verzeichn. formicid., 1863, p. 3; Dalla Torre,

Cat. Hymenop., 1893, 7, p. 221.
Camponotus ardens Mayr, Verb. Zool. bot. gesellsch. Wien, 1886, 36, p. 353.
Camponotus (Dinomyrmex) angusticollis Forel, Rev. Suisse zool., 1914, 22,

p. 268.

Type-locality: Malabar, Southern India (Jerdon).
Borneo (Mus. Caes.).

156. Camponotus (Myrmoturba) festinus (Smith).

Formica festina Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 53,

9.
Camponotus festinus Roger, Verzeich. formicid., 1863, p. 3; Dalla Torre, Cat.

Hymenop., 1893, 7, p. 231.
Camponotus [Myrmoturha) festinus Forel, Rev. Suisse zool., 1914, 22, p. 267.

Type-locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari; Bedot and Pictet).

AV heeler: the ants of BORNEO.

107

Two major workers and one minor from Kuching (John Hewitt),
one major from British North Borneo (E. B. Kershaw) and a minor
from Sadong, Sarawak (H. W. Smith) agree in measurements with the
type of this species as defined by Emery. Smith unfortunately based
the species on a female, so that, as Emery says, examination of the
type would probably give no clue as to whether it belonged to this
form or to the subspecies described by the Italian myrmecologist as
subsp. eximius from Sumatra.

*156a. Camponotus (Myrmoturba) festinus subsp. eximius

Emery.

Camponotus festinus subsp. eximius Emery, Ann. Mus. civ. Geneva, 1900,
ser. 2, 20, p. 701, ^ .
Type-locality: Si Rambe, Sumatra (E. Modigliani).
A single worker major from Kuching (John Hewitt).

*157. Camponotus (Myrmoturba) autrani Forel.

Camponotus autrani Forel, Ann. Soc. ent. Belgique, 1886, 30, p. 165, ^ .
Camponotus festinus subsp. autrani Emery, Ann. Mus. civ. Geneva, 1888,

ser. 2, 5, p. 528, ^ .
Camponotus {Myrmoturba) autrani Forel, Rev. Suisse zool., 1914, 22, p. 266.

Type-locality: Sumatra (Conrad Klaesi).

Two major workers from Sadong, Sarawak (H. W. Smith).

158. Camponotus (Myrmoturba) sucki Forel.

Camponotus sucki Forel, Mitth. Naturh. mus. Hamburg, 1901, 18, p. 71, ^ .
Camponotus (Myrmoturba) sucki Forel, Rev. Suisse zool., 1914, 22, p. 267.

Type-locality: Tandjong, S. E. Borneo (Fritz Suck).

159a. Camponotus (Myrmoturba) maculatus Fabricius subsp.

SETiTiBiA Forel.

Camponotus maculatus race setitibia Forel, Mitth. Naturh. mus. Hamburg,
1901, 18, p. 68, y .

Type-locality: Tandjong, S. E. Borneo (Fritz Suck).

108 bulletin: museum of comparative zoology.

159b. Camponotus (Myrmoturba) maculatus subsp. irritans

(Smith).

Formica irritans Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p.

55, ^.
Camponotus irritans Roger, Verzeichn. formicid., 1863, p. 3.
Camponotiis inconspicuus Mayr,'Ann. Mus. civ. Genova, 1872, 2, p. 135, ^ .
Camponotus (Myrmoturba) maculatus subsp. irritans Forel, Rev. Suisse zool.,

1914, 22, p. 267.

Type-locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari; Bedot and Pictet).
Numerous workers and a female from Kuching (H. W. Smith).

159c. Camponotus (Myrmoturba) maculatus subsp. irritans var.

inferior Forel.

Camponotus maculatus subsp. irritans var. inferior Forel, Rev. Suisse zool.,
1911, 19, p. 48, ^ ?. ,

Type-locality: Sarawak, Borneo (Haviland).
Several workers from Kuching (John Hewitt).

159d. Camponotus (Myrmoturba) maculatus subsp. compressus

(Fabricius).

Formica compressa Fabricius, Mant. Ins., 1787, 1, p. 307, ^ .

Formica indefessa Sykes, Trans. Ent. soc. London, 1835, 1, p. 104, § , pi. 13,

fig. 3.
Camponotus compressus Roger, Verzeichn. formicid., 1863, p. 2.
Camponotus quadrilaierus Roger, Berl. ent. zeitschr., 1863, 7, p. 136, ^ .
Camponotus maculatus subsp. compressus Emery, in Dalla Torre, Cat. Hy-

menop!, 1893, 7, p. 226.

Type-locality: Tranquebar (Hybner).
Sarawak (A. R. Wallace).

160. Camponotus (Myrmoturba) pallidus (Smith).

Formica pallida Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 57,

^.
Camponotus pallidus Mayr, Verh. Zool. bot. gesellsch. Wien, 1863, 13, p. 400;
Forel, Journ. Bombay nat. hist, soc, 1892, 7, p. 21, ^ ; Emery, Mem.
R. accad. sci. Bologna, 1896, 1895-1896, ser. 5, 5, p. 769.

wheeler: the ants of Borneo. 109

Type-locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari); Kapouas Basin (Chaper).
A single worker from Kuching (John Hewitt) from a colony " nest-
ing in a bunch of ratan fruits."

161. Camponotus (Myrmoturba?) fervens (Smith).

Formica fervens Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p.

55, ^.
Camponotus fervens Mayr, Verh. Zool. bot. gesellsch. Wien, 1886, 36, p. 354.
Camponotus (Myr?noturba?) fervens Forel, Rev. Suisse zool., 1914, 22, p. 267.

Type-locality: Sarawak, Borneo (A. R. Wallace).

162. Camponotus (Myrmotarsus) mistura (Smith).

Formica mistura Smith, Journ. Proc. Linn. soc. London, Zool., 1857, 2, p. 53, 9 .

Formica exasperata Smith, ibid., p. 56, S .

Camponotus exasperatus Mayr, Verh. Zool. bot. gesellsch. Wien, 1862, 12,

p. 659, ^ .
Camponotus mistura Mayr, ibid., 1886, 36, p. 354.
Camponotus {Myrmotarsus) mistura Forel, Rev. Suisse zool., 1914, 22, p. 269.

Type-locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari; Ha\aland).

A female from British North Borneo (E. B. Kershaw) and one from
Kuching (H. W. Smith).

163. Camponotus (Myrmotarsus) rufifemur Emery.

Camponotus rufifemur Emery, Ann. Mus. civ. Genova, 1900, ser. 2, 20, p.

705, ^ .
Camponotus (Myrmotarsus) rufifemur Torel, Rev. Suisse zool., 1914, 22, p. 269.

Type-locality: Mentawei (E. Modigliani).
Sarawak, Borneo.

A major worker from Kuching (John Hewitt) and a major and
minor worker from the Rambungan River (H. W. Smith).

164. Camponotus (Myrmotarsus) pressipes Emery.

Camponotus pressipes Emery, Ann. Soc. ent. France, 1893, p. 268 nota, 9 .
Camponotus (Myrmotarsus) pressipes Forel, Rev. Suisse zool., 1914, 22, p. 269.

110 bulletin: museum of comparative zoology.

Type-locality: Borneo.

Tandjong, S. E. Borneo (Fritz Suck).

Three females from Kuching (H. W. Smith).

165. Camponotus (Myrmotarsus) irritabilis (Smith).
Formica irritabilis Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 56, y .

Camponotus irritabilis Roger, Verzeichn. formicid., 1863, p. 3.

Camponotus (Myrmotarsus) irritabilis Forel, Rev. Suisse zool., 1914, 22, p. 269.

Type-locality: Sarawak, Borneo (A. R. Wallace).

Sarawak (Doria and Beccari).

Several workers from Kuching (John Hewitt) and the Rambungan
River, Sarawak (H. W. Smith) and a single worker transitional to the
var. sedulus from British North Borneo '(E. B. Kershaw).

165a. Camponotus (Myrmotarsus) irritabilis var. sedulus

(Smith).

Formica sedula Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 56, ^ .
Camponotus sedulus Mayr, Ann. Mus. civ. Genova, 1872, 2, p. 136, ^ 9 .
Camponotus irritabilis var. sedulus Forel, PlCV. Suisse zool., 1911, 19, p. 51.

Type-locality: Sarawak, Borneo (A. R. Wallace).
Four major workers from the Rambungan River, Sarawak (H. W.
Smith).

165b. Camponotus (Myrmotarsus) irritabilis var. winkleri

Forel.

Camponotus irritabilis var. winkleri Forel, Rev. Suisse zool., 1911, 19, p. 51, ^ .
Type-locality: Hayvep, Borneo (Winkler).

166. Camponotus (Myrmophyma) quadrisectus (Smith).

Formica quadrisecta Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 28, 9 .
Camponotus gilviceps Mayr, Tijdschr. ent. 1867, 10, p. 5, U .
Camponotus quadrisectus Forel, Ann. Soc. ent. Belgique, 1909, 53, p. 57 nota.
Camponotus {Myrmophyma) quadrisectus Forel, Rev. Suisse zool., 1914, 22,
p. 269.

wheeler: the ants of borneo. Ill

Type-locality: Philippine Islands.

Borneo (Ley den Mus.)-

A soldier and two minor workers from Baram, Borneo (John Hewitt)
taken from the " distorted pseudobulb of a Myrmecodia."

Forel assigns this species to his subgenus Myrmophyma, of which he
has designated it as the type, but it seems to me to belong more
naturally in Myrmotarsus, owing to the peculiar compression of the
hind metatarsi and the shape of the head. Emery had previously
(Mem. R. accad. sci. Bologna, 1896, ser. 5, 5, p. 773) placed the
species in his manipulus 11 of Camponotus with mistura Roger, platy-
pus Roger, etc.

167. Camponotus (Myrmophyma) exsectus Emery.

Camponotus exsectus Emery, Ann. Mus. civ. Geneva, 1900, ser. 2, 20, p. 709,
fig. 15, y 9 .

Type-locality: Sipora, Mentawei (E. Modigliani).
Pulo Laut, Borneo (Coll. Emery).

168. Camponotus (Myrmomalis) contractus Mayr.

Camponotus contractus Mayr, Ann. Mus. civ. Geneva, 1872, 2, p. 137, ^ .
Camponotus (Myrmomalis) contractus Forel, Rev. Suisse zool., 1914, 22, p. 271.

Type-locality: Sarawak, Borneo (Doria and Beccari).

169a. Camponotus (Myrmamblys) reticulatus Roger subsp.

BEDOTi Emery.

Camponotus hedoti Emery, Rev. Suisse zool., 1893, 1, p. 196, ^ , pi. 8, fig. 2.
Camponotus reticulatus subsp. yerburyi var. bedoti Emery, Mem. R. accad. sci.

Bologna, 1896, 1895-1896, ser. 5, 5, p. 772.
Camponotus {Myrmamblys) reticulatus subsp. bedoti Forel, Rev. Suisse zool.,

1914, 22, p. 271.

Type-locality: Batjan.
Sarawak (Bedot and Pictet).

170. Camponotus (Myrmosphincta) camelinus (Smith).
Formica camelina Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 57 y .

112 bulletin: museum of comparative zoology.

•

Camponotus senilis Mayr, Verb. Zool. bot. gesellsch. Wien, 1862, 12, p. 675, ^ ,

pi. 19, fig. 3.
Camponotus cavielinus Roger, Verzeichn. formicid., 1863, p. 3.
Camponohis singularis var. camelinus Emery, Ann. Mus. civ. Geneva, ser. 2, 4,

1887, p. 210.
Camponotus cinerascens var. camelinus Dalla Torre, Cat. Hymenop., 1893, 7,

p. 224.
Camponotus (Myrmosphinda) camelinus Forel, Mem. Soc. ent. Belgique, 1912,

20, p. 92.
Camponotus (Myrmocamelus) camelinus Forel, Rev. Suisse Zool., 1914, 22,

p. 270.

Type-locality: Singapore (A. R. Wallace).
Kapouas Basin, Borneo (Chaper).

Numerous workers from British North Borneo (E. B. Kershaw)
and Serambu Mt., Sarawak (H. W. Smith).

170a. Camponotus (Myrmosphincta) camelinus var. singularis

(Smith).

Formica singularis Smith, Cat. Hjonenop. Brit, mus., 1858, 6, p. 27, ^ .
Camponohis cinerascens Roger, Verzeich. formicid., 1863, p. 3; Dalla Torre,

Cat. Hymenop. 1893, 7, p. 224; Emery, Mem. R. accad. sci. Bologna,

1896, 1895-1896, ser. 5, 5, p. 771.
Camponotus singularis Mayr, Tijdschr. ent., 1867, 10, p. 39, ^ .

Type-locality: Java.
Sarawak (Doria and Beccari).

A worker from Matang Mt., W. Sarawak (H. W. Smith) and one
from Kuching (John Hewitt).

171. Camponotus (Myrmosphincta) dolichoderoides Forel.

Camponotus dolichoderoides Forel, Rev. Suisse zool., 1911, 19, p. 51, S .
Camponotus (Myrmosphincta) dolichoderoides Forel, Rev. Suisse zool., 1914,
22, p. 273.

Type-locality: Hayvep, Borneo (Winkler).

172. Camponotus (Myrmosphincta) hypoclineoides, sp. no v.
Worker minor. Length 5.3 mm.

wheeler: the ants of Borneo. 113

Slender; head nearly IJ times as long as broad, with the sides in front of
the moderately large and convex eyes, which are just anterior to the posterior
third of the head, rather straight and parallel, behind the eyes gradually
contracted to a narrow, marginate, but not colliform posterior border; in
profile very convex above and flat below. Mandibles narrow, feebly convex,
with straight external borders, their apical borders apparently 6- toothed.
Both the maxillary and labial palpi very long. Clypeus subcarinate behind,
its median portion trapezoidal, a little broader than long, its lateral portions
very narrow and reaching to the anterior corners of the head, its anterior
border straight and entire. Clypeal foveae small and deep. Frontal area
large, semicircular, rather indistinct. Frontal carinae sigmoidal, at their
greatest curvature fully as far apart as their distance from the lateral borders
of the head. Antennae long and slender, inserted well behind the clypeal
border, the nearly straight scapes extending about half their length beyond
the occipital border of the head. Thorax shaped like that of Hypoclinea,
broadest through the pronotum which, including the neck, is as long as broad,
evenly rounded and convex above ; mesonotum broadly and deeply constricted
behind, the anterior dorsal outline in profile falling rather steeply with an even
curve from the pronotum to the depression in which lie the two very prominent
metathoracic stigmata. Epinotum about as broad as the mesonotum, only a
little lower than the pronotum, with subequal base and declivity, the former
longitudinally and transversely rounded, rising rather steeply in front and
meeting the latter at a right angle which is neither sharp nor marginate.
Petiole narrow, as long as high, the node at its anterior end, erect and very
blunt and rounded, lower than the epinotum. Gaster elliptical. Legs long
and slender, distinctly compressed, tibiae without bristles on their flexor
surfaces.

Shining; mandibles subopaque, very finely and sparsely punctate; gaster
transversely alutaceous, with sparse, piligerous punctures.

Hairs long, erect, rather abundant, in part very long and flexuous on the
dorsal surface of the head, thorax, and petiole, whitish, with brown bases in
some lights, shorter on the front and sides of the head; long, abundant, and
oblique on the scapes and legs. Pubescence yellowish, confined to the an-
tennal funiculi.

Brown; fore tibiae and bases of middle and hind tibiae paler; neck, mandi-
bles, except the teeth, cheeks, anterior border of clypeus, antennal funiculus
except the basal half of the first joint, whitish yellow; posterior portion of
clypeus and antennal insertions somewhat darker; palpi dark brown, with
pale articulations.

Described from a single specimen taken on Mt. Matang, W. Sara-
wak by Prof. Harrison W. Smith.

This species is related to C. dolichoderoides Forel but is smaller,
shining and with very different pilosity and very differently shaped

114 bulletin: museum of comparative zoology.

head and petiole. It is also very different from C. horrens Forel of
the Philippines and moeschi Forel of Sumatra.

173. Camponotus (Myrmosphincta) megalonyx, sp. nov.

Worker major. Length about 7 mm.

Robust; head large, trapezoidal, broad behind and narrow in front, with
straight, anteriorly converging sides and deeply excised posterior border,
high and convex in the region of the vertex, with the posterior corners and
lateral borders feebly depressed. Eyes on the dorsal surface behind the
median transverse diameter of the head, rather small, moderately convex.
Mandibles stout and convex, their apical borders long, 6-toothed. Clypeus
rather flat, subcarinate behind, the median portion subhexagonal, as long as
broad, the lateral portions short, triangular, far from reaching the anterior
corners of the head; the anterior border sinuately excised on each side, with a
short, narrow, rounded lobe in the middle. Clypeal foveae pronounced.
Frontal area large, subtriangular; frontal groove very distinct. Frontal
carinae sinuous, diverging behind. Antennae short and stout; scapes feebly
curved, narrow, and terete at the base, gradually thickened towards their tips,
not reaching the posterior border of the head. Thorax very short; pronotum
broad, narrower than the head, convexly rounded, excluding the neck twice
as broad as long; mesonotura very short, broader than long, bordered anteri-
orly by a strong semicircular, promesonotal suture, rapidly sloping in profile
to a deep, narrow constriction only as broad as the prominent metanotal
stigmata on each side of it. Epinotum subcuboidal, only half as broad as the
pronotum, higher than long, with flat sides; in profile with the base abruptly
truncated in front. Straight and horizontal above and forming a right angle
with the flattened, truncated declivity which is narrowed below. Petiole
higher than long, subcuboidal, much lower than the epinotum, the node thick,
higher than long, the anterior surface slightly concave and higher than the
posterior surface, the dorsal surface rounded and sloping backward and down-
ward. Gaster broadly elliptical, smaller than the head, somewhat flattened.
Legs stout, the terminal tarsal joint, claws, and empodia noticeably enlarged;
tibiae slightly flattened but not chanelled, without a row of bristles on their
flexor surfaces.

Opaque ; head and gaster slightly lustrous ; very finely, densely, and evenly
punctate, with sparser piligerous punctures, abundant and transverse on the
gaster. Mandibles and legs shining, very finely and superficially shagreened,
the mandibles also coarsely and sparsely punctate and near their apical margins
coarsely striate; legs with sparse piligerous punctures.

Hairs reddish, rather abundant, erect; long and flexuous on the thoracic
dorsum and first gastric segment, shorter elsewhere; absent on the cheeks and
almost absent on the antennal scapes, present on the gula and legs. Pubescence
long and abundant but nowhere concealing the sculpture, conspicuous on the
posterior portion of the head, the thorax and gaster.

wheeler: the ants of borneo. 115

Black; thorax and petiole dark brown; upper surface of epinotum and some
spots on the pleurae ferruginous; coxae and femora brownish yellow; tibiae,
tarsi, and funiculi dark brown; terminal tarsal joints, claws, and empodia
reddish.

Worker minor. Length nearly 5 mm.

Very similar to the major worker, except that the head is much smaller,
proportionally longer, with straight posterior border, more prominent eyes
and the antenna] scapes extending about -g- their length beyond the occipital
border. The epinotal declivity is slightly concave and a little more sloping,
the petiolar node is decidedly thinner. The thorax and legs are dark brown,
except the articulations of the latter and the tarsi beyond the basal joint,
which are reddish. The erect hairs on the legs are somewhat less numerous.

Described from two major and two minor workers taken by Mr.
William Beebe on the Mujong River, Sarawak, "running on bushes."

Like the preceding, this is a very peculiar species, which I have
placed in Forel's subgenus Myrmosphincta on account of its thoracic
structure. In my opinion this subgenus is an unnatural assemblage
of forms, but in the present stage of myrmecology it is useful as a
catch-all for the species Avith markedly sellate thorax. The tarsi and
claws of C. mcgalonyx show that it is a true arboreal ant. It exhibits
certain peculiarities in the structure of the clypeus and thorax that
recall the conditions in Colobopsis.

174. Camponotus (Colobopsis) pilosus (Smith).

Forrnica pilosa Smith, Journ. Proc. Linn. soc. London. ZooL, 1857, 2, p. ,54, ^ .
Colobopsis ptibescens Mayr, Verh. Zool. bot. gesellsch. Wien, 1862, 12, p. 691,

S 9 («ec Fabricius).
Camponotus pubescens Emery, Ann. Mus. civ. Genova, 1889, ser. 2, 7, p. 517.
Camponohis (Colobopsis) leonardi Emery, ibid., p. 515, 9 .
Camponotus {Colobopsis) pilosus Forel, Rev. Suisse zool., 1914, 22, p. 272.

Type-locality: Sarawak, Borneo (A. R. Wallace).
Kapouas Basin (Chaper).

Soldiers, workers, and females from British North Borneo (E. B.
Kershaw), Kuching and Rambungan River (H. W. Smith).

175. Camponotus (Colobopsis) badius (Smith).

Formica badia Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 54, U .
Camponotus badius Mayr, Verh. Zool. bot. gesellsch. Wien, 1886, 36, p. 354.
Camponotus (Colobopsis) badius Forel, Rev. Suisse zool., 1914, 22, p. 272.

116 bulletin: museum of comparative zoology.

Type-locality: Singapore (A. R. Wallace).
Sarawak (A. R. Wallace).

A worker from Kuching and one from the Rambungan River (H. W.
Smith).

176. Camponotus (Colobopsis) clerodendri Emery.

Colobopsis clerodendri Emery, Ann. Mus. civ. Geneva, 1887, ser. 2, 4, p. 241,

01 S , fig.
Camponotus (Colobopsis) clerodendri Emery, ibid, 1889, ser. 2, 7, p. 517; Forel,

Rev. Suisse zool., 1914, 22, p. 272.
Camponotus clerodendri Dalla Torre, Cat. Hymenop., 189.3, 7, p. 225.

Type-locality: Sarawak, Borneo (Doria and Beccari).

Two soldiers and two workers from Kuching (John He^vitt). A note
accompanjdng the specimens states that "this ant vomits a yellow,
sticky juice."

177. Camponotus (Colobopsis) doriae Mayr.

Camponotus doriae Mayr, Ann. Mus. civ. Geneva, 1872, 2, p. 137, ^ .
Camponotus (Colobopsis) doriae Forel, Rev. Suisse zool., 1914, 22, p. 272.

Type-locality: Sarawak, Borneo (Doria and Beccari).
Two workers from Kuching (John Hewitt).

178. Camponotus (Colobopsis) fasciatus Mayr.

Colobopsis fasdata Mayr, Tijdschr. ent., 1867, 10, p. 57, 9 .
Camponotus (Colobopsis) fasciatus Emery, Ann. Mus. civ. Genova, 1889, ser.
2, 7, p. 517.

Type-locality: Java (Leyden Museum).
Kapouas Basin, Borneo (Chaper).

*179. Camponotus (Colobopsis) saundersi Emery.

Camponotus (Colobopsis) saundersi Emery, Ann. Mus. civ. Genova, 1889,
ser. 2, 7, p. 516, g .

Type-locality: Thagata, Tenasserim (L. Fea).

Six minor workers from the Sarawak River, near Kuching (H. W.
Smith).

wheeler: the ants of borneo. 117

*180. Camponotus (Colobopsis) vitreus (Smith).

Formica vitrea Smith, Journ. Proc. Linn. soc. London. Zool., 1861, 5, p. 94, ^
Camponotus vitreris Emery, in Dalla Torre, Cat. Hymenop., 1893, 7, p. 257.
Camponotus (Colobopsis) vitreus Forel, Rev. Suisse zool., 1914, 22.

Type-locality: Batjan (A. R. Wallace).

A single worker from British North Borneo (E. B. Kershaw).

181. Camponotus (Colobopsis) strictus (Jerdon).

Formica stricta Jerdon, Madras Journ. lit. sci., 1851, 17, p. 123, ^ .
Colobopsis stricta Mayr, Verb. Zool. bot. gesellsch. Wien, 1886, 36, p. 353.
Camponotus strictus Dalla Torre, Cat. Hymenop. 1893, 7, p. 253.
Camponotus (Colobopsis) strictus Emery, Ann. Mus. civ. Genova, 1889, ser. 2,
7, p. 517.

Type-locality: Malabar, Southern India (Jerdon).
Sarawak (A. R. Wallace).

182. Camponotus (Colobopsis) vigilans (Smith).

Formica vigilans Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 26, S .
Colobopsis vigilans Mayr, Verh. Zool. bot. gesellsch. Wien, 1886, 36, p. 354.
Camponotus (Colobopsis) vigilans Emery, Ann. Mus. civ. Genova, 1889, ser. 2,

7, p. 517.
Camponotus vigilans Dalla Torre, Cat. Hymenop., 1893, 7, p. 257.

Type-locality: Sarawak, Borneo.

*183. Camponotus (Colobopsis) leucodiscus, sp. nov.

FemUle (dealated). Length 5 mm.

Head subrectangular, longer than broad, broader behind than in front, with
straight posterior and lateral borders, indistinctly truncate in front and convex
dorsally. Eyes large, behind the middle of the head, separated by a distance
equal to their length from the anterior corners of the head. Mandibles con-
vex, with 5 subequal teeth. Clypeus flat, ecarinate, subrectangular, a little
longer than broad, as broad behind as in front, its lateral borders notched just
in front of the middle at the deep, pit-like clyijeal foveae, the anterior border
rounded and entire. Frontal area lozenge-shaped; frontal groove distinct

118 bulletin: museum of comparative zoology.

but not reaching the anterior ocellus. Frontal carinae sinuous, diverging
behind. Antennae inserted some distance behind the clypeus; the scapes
terete, slightly thickened at their tips, reaching a little beyond the posterior
corners of the head; funicular joints all very distinctly longer than broad.
Thorax very regularly elongate-elliptical, nearly 2^ times as long as broad, as
broad as the head. Mesonotum convex and rounded in front, flattened be-
hind, as long as broad. Epinotum with distinct base and declivity, the former
about half as long as the latter, meeting it at a rounded right angle. Petiole
low and small, the node very blunt and rounded, about twice as broad as long.
Gaster regularly elliptical, a little shorter and broader than the thorax. IjCgs
rather short.

Subopaque; head in front and especially the mandibles more shining;
mandibles and clj'peus very finely and sparsely punctate; remainder of the
body very finely and densely punctate, with coarser, sparse, piligerous punc-
tures, most distinct on the front and posterior portion of the head.

Head, thora.x, and gaster with very sparse, delicate, erect, blunt, blackish
hairs; antennae and legs with short, sparse, appressed, rather indistinct, pale
pubescence.

Black or very dark brown; mandibles, except their teeth, clypeus, cheeks,
sides of front, antennae and their insertions, reddish yellow. Gaster milk-
white, dorsal surface black, with a large, median, irregularly elliptical white
spot, extending from the anterior third of the first to the posterior border of
the fourth segment; the posterolateral margins of the second to fourth seg-
ments also white; anal segment black, with yellowish tip. Wing-insertions,
coxae, and trochanters white; tarsi, anterior surface of the fore legs and ventral
portions of the fore femora brownish yellow.

Described from a single specimen taken by Mr. E. B. Kershaw in
British North Borneo.

This species is readily distinguished by its unusual coloration from
any of the Colobopsis of which I have seen specimens or descriptions.
Its small size indicates that the workers must be diminutive, like those
of the European and North American species.^

184. Camponotus (Colobopsis) gilviceps Roger.

Formica ruficeps Smith, Journ. Proc. Linn. soc. London. Zool., 18-57, 2, p. 54,

y {nee Fabricius).
Camponotus gilviceps Roger, Verzeichn. formicid., 1863, p. 3; Forel, Ann. Soc.

ent. Belgique, 1909, 53, p. 57 nota.

1 Since the foregoing description was written I have received from Dr. J. W. Chapman a
fine series of the soldiers and workers of this ant taken at Dumaguete, Negros Oriental, in The
Philippines. The soldiers are colored like the female, but in the worker the white gastric spot
is shorter, extending only to or slightly beyond the posterior margin of the second segment.

wheeler: the ants of borneo. 119

Camponotus {Colobopsis) gilviceps Forel, Rev. Suisse zooL, 1914, 22, p. 272.
Type-locality: Sarawak, Borneo (A. R. Wallace).

185. Camponotus (Colobopsis) hosei Forel.

Camponotus hosei Forel, Rev. Suisse zool., 1911, 19, p. 55, ^ .
Camponotus (Colobopsis) hosei Forel, ibid., 1914, 22, p. 272.

Type-locality: Borneo (Hose).

186. Camponotus (Colobopsis) hosei var. mimus Forel.

Camponotus hosei var. mimus Forel, Rev. Suisse zool., 1911, 19, p. 56, 9 .
Type-locality: Borneo (Hose).

187. Camponotus (Colobopsis) severini Forel.

Camponotus (Colobopsis) severini Forel, Ann. Soc. ent. Belgique, 1909, 53,

p. 55, Ql g 9 o".

Type-locality: Labuan Island, Borneo.

188. Camponotus (Colobopsis) smithianus, sp. nov.

Worker major. Length 5-5.6 mm.

Head large, regularly oblong, 1^ times a^ long as broad, equally flattened
above and below, obliquely truncated in the clypeal region, with the cheeks
swollen and rounded and their borders projecting anteriorly around the in-
sertions of the mandibles as curved plates ; the region in front of the eyes and
near the frontal carinae distinctly impressed on each side ; the eyes rather small
and flat, their anterior orbits just in front of the posterior third of the head.
Mandibles very stout and convex, with five obtuse teeth. Palpi very short.
Clypeus with distinct median and lateral portions, the former ecarinate, con-
sisting of two planes, a posterior continuing the dorsal surface of the head,
rectangular and nearly twice as broad as long, and an anterior, obliquely
sloping and semicircular, its median curved edge forming the anterior border
of the clypeus. The triangular, scale-like lateral portions of the clypeus are
marked off by grooves that terminate behind in the small but distinct cljrpeal
foveae. Frontal area indistinct; frontal groove delicate; frontal carinae long,
lyriform, continued nearly as far back as the posterior orbits, diverging behind

120 bulletin: museum of comparative zoology.

where they are twice as far apart as their distance from the lateral corners of
the head. Antennae short and slender, the scapes uniformly bent, slightly
flattened though narrow at the base, gradually enlarging towards their tips,
which reach the posterior corners of the head; funiculi of very uniform thick-
ness throughout, the joints subequal, a little longer than broad, the first and
last longer. Thorax much narrower and a little shorter than the head, the
pro- and mesonotum together forming a hemispherical mass, with strong
mesoepinotal suture, the mesonotum broader than long. Mesoepinotal
constriction rather deep and very short; metanotal sclerite distinct, short, and
convex; epinotum very small, its base long, flattened, nearly perpendicular,
the declivity extremely short and strongly concave. Petiole thick, and very
low; from above twice as broad as long and a little narrower in front than
behind, in profile truncated anteriorly, posteriorly, and dorsally, the posterior
surface higher than the anterior and distinctly impressed in the middle above.
Gaster elongate-elliptical, smaller than the head. Legs rather stout, slightly
flattened, the fore femora enlarged, the claws well developed.

Mandibles shining, finely and evenly but not densely punctate; anterior
half of head subopaque, densely and finely punctate and minutely and in-
distinctly rugulose; remainder of body shining, distinctly shagreened, the
gaster transversely, covered with sparse, piligerous punctures.

Hairs grayish, coarse, erect, of uneven length, longest and moderately abund-
ant on the dorsal surface of the head, thorax, petiole, and gaster; absent on
the gula; short, even, and oblique on the legs and distal portions of the antennal
scapes. Pubescence pale, long, coarse, and sparse, conspicuous on the sides
of the head, between the frontal carinae and on the gaster.

Black; mandibles, cheeks, and clypeus cherry-red; antennal funiculi yellow,
except the first and last joint, which are black or piceous like the scapes;
tarsi and articulations of the thorax, petiole, and legs brown or reddish. ,

Worker minor. Length 3.5 mm.

Head subtrapezoidal, as broad' as long, a little narrower in front than behind,
with feebly convex sides and posterior border, only moderately and evenly
convex above, with the eyes at the posterior third. Mandibles narrow, with
oblique, apparently 4-toothed apical and straight lateral borders. Maxillary
palpi much longer than in the major worker. Clypeus broader than long,
subtrapezoidal, feebly carinate, with nearly straight, entire anterior border.
Frontal area obsolete; frontal groove feeble; frontal carinae short, rather
straight, strongly diverging behind. Antennae long, the scapes extending
about I their length bej-ond the posterior corners of the head; first funicular
joint nearly as long as the second and third together, the second distinctly
shorter than the third, which like the fourth and fifth is nearly twice as long
as broad. Thorax in profile composed of two subequal hemispherical masses,
one formed bj' the pro- and mesonotum, the other by the meta- and epinotum,
meeting at a short, acute, mesoepinotal incision; the epinotum really without

wheeler: the ants of borneo. 121

distinct base and declivity, though the lower metasternal border is reflected
at the articulation with the petiole. Petiole resembling that of the major
worker, but even lower, with its dorsal surface more truncated and flattened.

Sculpture and pilosity much as in the worker major, but the anterior portion
of the head not subopaque and punctate, the pubescence on the sides and
front of the head replaced by oblique or subappressed, rather abundant hairs
and the scapes as abundantly provided with oblique hairs as the legs.

Black; mandibles, anterior border of cheeks and clypeus, the mesopleurae,
sides and posteroinferior border of the epinotum, and the tarsi beyond the first
joint, red. Funicular joints 3-9 brownish yellow.

Described from three major workers and one minor worker taken on
the Rambungan River, Sarawak by Prof. Harrison W. Smith, to whom
the species is dedicated. It is very distinct from all the Indomalayan
Colobopsis of which I have seen specimens or descriptions. Type.
M. C . Z. 9,074.

189. Camponotus (incertae sedis) tenuipes (Smith).

Formica tenuipes Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 57, 9.
Camponotus tenuipes Mayr, Verh. Zool. bot. gesellsch. Wien, 1883, 36, p. 354.

Type-locality: Sarawak, Borneo (A. R. Wallace).

190. POLYRHACHIS (POLYRHACHIs) BIHAMATA (Drury).

Formica bihamata Drury, Illustr. nat. hist., 1773, 2, pi. 38, fig. 7, 8, ^ .
Formica affinis Le Guillou, Ann. Soc. ent. France, 1841, 10, p. 314, 9 .
Polyrhachis bihamata Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 58, pi. 1, fig. 9.
Polyrhachis affinis MajT, Verh. Zool. bot. gesellsch. Wien, 1863, 13, p. 443.

Type-locality: Johanna Island, Comoro Islands.
Sarawak, (Doria and Beccari; A. R. Wallace); Kapouas Basin
(Chaper).

Nmnerous workers from Sarawak River, Kuching, Serambu, Sara-
wak (H. W. Smith) and Kuching (John HeT\dtt).

191. Polyrhachis (Polyrhachis) bellicosa Smith.

Polyrhachis bellicosus Smith, Journ. Proc. Linn. soc. London. Zool., 1859, 3 ,
p. 142, ^ .

122 bulletin: museum of comparative zoology.

Polyrhachis hihamata var. bellicosa Mayr, Verb. Zool. bot. gesellsch. Wien,

1862, 12, p. 677.
Polyrhachis bellicosa Mayr, Tijdschr. ent. 1867, 10, p. 50, ^ .

Type-locality: Am (A. R. Wallace).

Sarawak (Doria and Beccari; Bedot and Pictet).

Numerous workers from Sadong, Serambu, and Sarawak River,
Kuehing (H. W. Smith), British North Borneo (E. B. Kershaw), and
Kuehing (John Hewitt).

192. Polyrhachis (Polyrhachis) ypsilon Emery.

Polyrhachis bihamata Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 58, S .
Polyrhachis ypsilon Emery, Ann. Mus. civ. Geneva, 1887, ser. 2, 4, p. 239, ^ .

Type-locality: Borneo.

Fourteen workers from Sadong and Kuehing (H. W. Smith) .

193. Polyrhachis (Campomyrma) equina Smith.

Polyrhachis equinus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 63, S-. »

Polyrhachis equina Dalla Torre, Cat. Hymenop., 1893, 7, p. 261.
Polyrhachis biloba Forel, Rev. Suisse zool., 1911, 19, p. 58, ^ .

Type-locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Haviland).

Examination of a single specimen taken by John Hewitt at Kuehing
shows that Forel has redescribed Smith's Polyrhachis equina as biloba.

194. Polyrhachis (Campomyrma) gribodoi Emery.

Polyrhachis gribodoi Emery, Ann. Mus. civ. Genova, 1887, ser. 2, 4, p. 22, ^ 9 .

Type-locality: Java.
Sarawak (Doria and Beccari).

195. Polyrhachis (Myrma) relucens (Latreille).

Formica relucens Latreille, Hist. nat. fourmis., 1802, p. 131, ^ , pi. 4, fig. 24.
Polyrhachis relucens Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 59.

wheelek: the ants of Borneo. 123

Type-locality: East Indies fRiche and La Billardiere).
Sarawak (A. R. Wallace).

196. PoLYRHACHis (Myrma) pruinosa MayT.

PolyrhacMs pruinosa Mayr, Ann. Mus. civ. Geneva, 1872, 2, p. 142, ^ .
Type-locality: Sarawak, Borneo (Doria and Beccari).

197. PoLYRHACHis (Myrma) murina Emery.

Polyrhachis murina Emery, Rev. Suisse zool., 1893, 1, p. 198, ^ .
Type-locality: Sarawak, Borneo (Bedot and Pictet).

198. Polyrhachis (Myrma) mayri Roger.

Polyrhachis rehicens Mayr, Verh. Zool. bot. gesellsch. Wien, 1862, 12, p. 685,

y (nee Latreille).
Polyrhachis maijri Roger, Verzeich. formicid., 1863, p. 7; Mayr, Tijdschr.

ent. 1867, 10, p. 56, U .

Type-locality: Java (Kirsch).

Sarawak (Doria and Beccari); Kapouas Basin (Chaper).

Two workers from Serambu Mt., Sarawak (H. W. Smith).

199. Polyrhachis (Myrma) lycidas Smith.

Polyrhachis lycidas Smith, Journ. Proc. Linn. soc. London. Zool., 1861, 6,
p. 43, y , pi. 1, fig. 23; Mayr, Tijdschr. ent., 1867, 10, p. 31, S 9 .

Type-locality: Celebes.

Borneo fLevden Mus.).

Two Avorkers from Kuching (John Hewitt).

200. Polyrhachis (Myrma) beccarii Mayr.

Polyrhachis beccarii MajT, Ann. Mus. civ. Genova, 1S72, 2, p. 141, ^ ; Emery,
ibid., 1887, ser. 2, 4, p. 232, ^ .

Type-locality: Sarawak, Borneo (Doria and Beccari).
Two workers from Kuching (John Hewitt).

124 bulletin: museum of compakative zoology.

201. PoLYRHACHis (Myrma) nigropilosa MayT.

Polyrhachis nigropilosa Mayr, Ann. Mus. civ. Geneva, 1872, 2, p. 141, ^ ;
Emery, ibid, 1887, ser. 2, 4, p. 232, ^ .

Type-locality: Sarawak, Borneo (Doria and Beccari).
Three workers from Serambu Mt., Sarawak and Kuching (H. W.
Smith) and British North Borneo (E. B. Kershaw).

201a. Polyrhachis (Myrma) nigropilosa var. conophthalma

Emery.

Polyrhachis nigropilosa var. conophthalma Emery, Ann. Mus. civ. Geneva,
1900, ser. 2, 20, p. 713, ^ .

Type-locality: Smnatra (E. Modighani).

A single worker from Serambu Mt., Sarawak (H. W. Smith).

202. Polyrhachis (Myrma) sculpturata Smith.

Polyrhachis sculpturatus Smith, Journ. Proc. Linn. soc. London. ZooL, 1861,

5, p. 70, ^ 9 .
Polyrhachis sculpturata Mayr, Tijdschr. ent., 1867, 10, p. 59, y .

Type-locality: Makassar, Celebes (A. R. Wallace).
Kapouas Basin, Borneo (Chaper).

203. Polyrhachis (Myrma) striata Mayr.

Polyrhachis striatum Mayr, Verh. Zool. hot. gesellsch. Wien, 1862, 12, p. 686, ^ ,

pi. 19, fig. 8.
Polyrhachis striata Mayr, Reise Novara. ZooL, 2, Formicid., 1865, p. 44, ^ ,

pi. 2, fig. 11.

Type-locality: Java (Novara Expedition).

Sarawak (Doria and Beccari).

A single worker from Kuching (John Hewitt).

204. Polyrhachis (Myrma) sumatrensis Smith.

Polyrhachis sumatrensis Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 65, 9 ,
pi. 4, fig. 43; Forel, Sitzb. K. bayr. akad. wiss. Math. phys. klasse, 1911,
p. 296-.

wheeler: the ants of borneo. 125

Type-locality: Sumatra.
Sarawak (Munich Mus.)-

205. PoLYRHACHis (Myrma) villipes Smith.

PolyrhacMs villi'pes Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 61, ^ ; Mayr, Tijdschr. ent., 1867, 10, p. 58, ^ .

Type-locality: Sarawak, Borneo (A. R. Wallace).
Sarawak (Doria and Beccari; Haviland).

Two workers and a female from Kuching (John Hewitt) and a
worker from the same locality (H. W. Smith).

206. PoLYRHACHis (Myrma) vindex Smith,
Polyrhachis vindex Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p.

64, y .

Type-locality: Sarawak, Borneo (A. R. Wallace).
Worker. Length: 5-6 mm.

Head broadly oval, a little longer than broad, and a little broader behind
than in front, nearly as high in the region of the frontal carinae as long. Eyes
distinctly behind the middle of the sides, moderately large and convex. Mandi-
bles with feebly convex external and oblique, coarsely 4-toothed apical border.
Clypeus subcarinate, its anterior border feebly and sinuately emarginate in
the middle and on each side. Frontal area distinct, triangular; frontal
carinae closely approximated, erect, very close together anteriorly. Antennae
long; scapes slender, reaching about f their length beyond the occiput.
Thorax short, only a little longer than its greatest height, the sides flat, the
dorsum evenly convex and rounded, its sides acutely marginate, the margin
incised at the distinct promesonotal and at the nearly obsolete mesoepinotal
suture. Pronotum, excluding the neck, nearly twice as broad as long, its
anterior corners produced as flattened, acute spines which are nearly twice as
long as broad at their bases, and directed forward and very slightly outward.
Mesonotum transversely oblong, nearly as broad as the pronotum and nearly
four times as broad as long. Base of epinotum trapezoidal, somewhat less than
twice as broad as long, with straight sides and posterior border, the latter
strongly marginate, with a small, acute, upturned tooth at each corner, the
declivity abrupt, concave and narrowed below, as long as the base and margi-
nate on the sides. Petiole lenticular, as broad as the anterior border of the
epinotum, in profile convex anteriorly and posteriorly, with a sharp superior
border, which seen from behind is evenly arcuate and entire, with a small,

126 bulletin: museum of comparative zoology.

rather acute, triangular tooth on each side. Below the teeth the sides are
straight and rapidly converge. Gaster subglobose, with very large first seg-
ment, strongly and concavely truncated in front. Legs long and stout, tibiae
cylindrical, constricted at the base, without bristles on their flexor surfaces.

Subopaque; mandibles finely and evenly striated, with minute, scattered
punctures; clypeus finely and densely punctate; upper surface of head,
dorsvun and sides of thorax evenly and rather finely longitudinally rugose with
indistinctly punctate interrugal spaces; sides of head finely and somewhat
irregularly rugulose-punctate; gula smooth and shining; declivity of epinotum
finely, transversely rugulose; petiole and gaster lustrous, very finely and
indistinctly punctate.

Hairs very sparse, whitish, erect, confined to the head and tips of the gaster.
Pubescence whitish, indistinct on the head; longer and more abundant on the
pleurae; very fine and dense, producing a pruinose appearance on the thoracic
•dorsum, petiole, and gaster, less apparent on the legs.

Black; palpi reddish; legs variable in color; in some specimens black
throughout, in others with the femora and tibiae red or reddish yellow and the
knees, coxae, and tarsi black.

Eight workers ; three from Kuehing (John Hemtt " nesting at base
of an epiphyte between intertwining roots"), one from Kuehing and
one from Serambu Mt. (H. W. Smith), two from Bongo Mt. (Hewitt
and Brooks) and one from Mt. Matang, West Sarawak (G. E. Bryant).

I have redescribed this form in detail because it seems certain that
it is Smith's P. xindex, which has not been recognized up to the present
time. The species is evidently very close to Smith's inermis and
■orsyllvs, especially to the latter, but both of these forms were described
later and would therefore be only subspecies or synonyms of vindex.

*207a. PoLYRHACHis (Hemioptica) aculeata Ma;yT subsp. cybele,

subsp. nov.

Worker. Length 6-6.5 mm.

Differing from the typical form in its somewhat larger size, broader head, and
decidedly larger and higher petiole, with its two spines a little shorter and the
transverse margin between them more arcuate and sharper. The eyes are a
little larger and distinctly less truncated laterally and less conical and more
hemispherical when seen from the front. The erect, pale hairs on the body
are distinctly less abundant than in the typical aculeata and the legs are en-
tirely black.

Described from four specimens from Kuehing (John Hewitt).
Type—M. C. Z. 9,076.

WHEELEK: the ants of BORNEO. 127

*208a. PoLYRHACHis (Myrmothrinax) thrinax Roger var.

JAVANICA Mayr.

Polyrhachis thrinax var. javanica Mayr, Tijdschr. ent., 1867, 10, p. 20, ^ .

Type-locality: Java (Mus. Holm.).

Three workers from Kuching (John Hewitt).

209. Polyrhachis (Myrmothrinax) aequicuspis, nora. nov.

Polyrhachis constructor Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 68, 9 ,
pi. 4, fig. 23 {:nec P. constructor Smith, Journ. Proc. Linn. soc. London.
Zool., 1857, 2, p. 60).

Type-locality: Sarawak, Borneo (Saunders Coll.).

It seems not to have been noticed that Smith gave the name con-
structor to two very different species of Polyrhachis. In 1857 he
described a P. construcior from a female specimen taken by A. R.
Wallace at Sarawak. This specimen measured 3| lines and had the
petiole "quadrate, with two very stout, short, curved spines above"
and the body covered with "short, silky, ashy pile, most apparent
on the abdomen." In 1858 he gave the same name to a female from
the same locality but from the W. W. Saunders Collection, measuring
4| lines, with the " node of the petiole incrassate and with three stout,
short, acute spines." The figure accompanying this description
shows that he had befqj'e him a species of Myrmothrinax, with sub-
equal petiolar spines, possibly the form afterwards called thrinax var.
jaxanica by Mayr. It therefore becomes necessary to change the
name of the second constructor. The first ■wall have to be consigned
to the species of Polyrhachis inccrtae sedis for the present {vide infra
p. 137).

*210. Polyrhachis (Myrmothrinax) triaena, sp. nov.

Worker. Length 6.5-7 mm.

Head a little longer than broad, a little broader behind than in front, with
evenly rounded and convex occipital border and straight sides; very convex
dorsally in the frontal region. Eyes large and convex, broadly elliptical,
their anterior orbits at the median transverse diameter of the head. Mandi-
bles very convex, with five large, subequal teeth. Ch-peus carinate, broader

128 bulletin: museum of comparative zoology.

than long, its anterior border entire, evenly rounded and slightly projecting.
Frontal area distinct, triangular; frontal groove narrow but distinct; frontal
carinae rather approximated, prominent, sinuous, as near together behind
as in front. Antennae long; scapes extending more than half their length
beyond the occipital border, the basal funicular joints about 2i times as long
as broad. Thorax slender, its dorsal and lateral surfaces flattened, the former
strongly marginate on all sides, except at the neck. Promesonotal and meso-
epinotal sutures distinct, but not incised where they meet the lateral margina-
tions. Dorsum of pronotum as broad as long, with two flat, blunt teeth
anteriorly, as long as broad at their bases, directed outwards; dorsum of
mesonotum trapezoidal, about li times as long as broad, with straight sides,
converging posteriorly. Base of epinotum regularly oblong, twice as long as
broad, its posterior angles forming two small erect teeth, as large as those on
the pronotum but more acute; epinotal declivity shorter than the base, sloping
and concave. Petiole from above U times as long as broad, broader in front
than behind, bearing above in the middle a long, straight, backwardly directed
spine, nearly as long as the greatest height of the remainder of the petiole, and
on each side a triangular tooth, only a little longer than broad at the base and
somewhat blunter than the median spine. Gaster a little larger than the head,
oval, broad in front. Legs long; tibiae cylindrical, distinctly constricted at
the base, without a row of bristles on their flexor surfaces.

Mandibles slightly shining, very finely striated and finely and sparsely
punctate; clypeus, head, thorax, and petiole densely and evenly punctate.
Gaster and legs very finely and superficially s-hagreened, the former shining,
the latter lustrous.

Hairs very few, short, obtuse, yellowish, confined to the clypeus, mandibles,
and tip of the gaster; pubescence extremely fine and dilute, visible only on
the antennae and tibiae.

Brownish ferruginous; scapes, legs, and gaster a Uttle paler than the head,
thorax, and petiole.

Female (deiilated). Length nearly 10 mm.

Eesembling the worker in the shape of the head and petiole. Thorax
elongate-elliptical, nearly 2-i times as long as broad, narrower than the head;
meso- and epinotum not marginate on the sides, the former as broad as long,
convex in front, flattened behind; the latter with subequal base and declivity,
the base slightly convex and not separated from the sloping, concave declivity
by a transverse ridge as in the worker; epinotal teeth stout, short, and blunt,
directed upward. Pronotum with the blunt, triangular teeth at the anterior
corners not connected by a distinct transverse carina. Median spine of the
petiole of rather uniform thickness, with blunt tip.

Sculpture resembling that of the worker, but the head, thorax, and petiole
delicately reticulate-rugose as well as finely punctate.

Black ; tip of last funicular joint yellowish ; gaster with a shghtly reddish tint.

wheeler: the ants of Borneo. 129

Described from two workers and a female taken by Mr. John
Hewitt at Kuching.

Of the various species of Myrmothrinax, namely thrinax Roger,
textor Smith, aequicuspis Wheeler, dahli Forel and frauenfeldi Mayr, this
species seems to be most closely related to the last. The worker
triacna, however, is smaller, not black and has the sides of the thorax
sharply marginate, whereas Mayr says of frauenfeldi that it has the
" thorax sine marginibus aciitis." The Bornean form is not unlike
dahli in the general shape of the thorax, but the pronotal teeth of the
former are longer, the epinotal teeth much shorter, the sculpture,
color, and size are different and the basal funicular joints are shorter.
P. friaena occurs also in Java as I possess a dealated female from that
island received from Staudinger under the name frauenfeldi.

211. PoLYRHACHis (Chariomyrma) arcuata (Le Guillou).

Formica arcuata Le Guillou, Ann. Soc. ent. France, 1841, 10, p. 315, ^ 9 .
Polyrhachis latifrons Roger, Berlin ent. zeitschr. 1863, 7, p. 155, ^ .
Polyrhachis modiglianii Emery, Ann. Mus. civ. Geneva, 1888, ser. 2, 5, p. 529,

y 9, pi. 9, fig. 1.
Polyrhachis arcuata Forel, Mitth. Zool. mus. Berlin, 1901, 2, p. 32.

Type-locality: Borneo (Voyage of the "x\strolabe" and "Zelee").
A single dealated female from Bongo Mt., Sarawak (Hewitt and
Brooks).

212. Polyrhachis (Myrmhopla) armata (Le Guillou).

Formica armata Le Guillou, Ann. Soc. ent. France, 1841, 10, p. 313, cf.

Polyrhachis armata Mayr, Tijdschr. ent., 1867, 10, p. 46, ^ .

Polyrhachis defensus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 59, y .

Polyrhachis pandarus Smith, ibid, p. 62, ^ .

Tyj)c-locality: Samboangan, Philippines (Voyage of the "Astro-
labe" and "Zelee").

Sarawak (Doria and Beccari; A. R. Wallace); Hayvep (Winkler).

Numerous workers and females from Kuching (John Hewitt) and
Kuching, Matang, and Serambu Mts. (H. W. Smith) and British
North Borneo (E. B. Kershaw). Nearly all of these specimens have
the gaster ferruginous instead of black.

130 bulletin: museum of comparative zoology.

213. Polyrhachis (Myrmhopla) lugens Mayr.
Polyrhachis lugens MajT, Tijdschr. ent., 1867, 10, p. 31, 9 .
Type-locality: Borneo (Ley den Museum).

*214. Polyrhachis (Myrmhopla) pressa Mayr.

Polyrhachis pressus Mayr, Verb. Zool. bot. gesellsch. Wien, 1862, 12, p. 681,

^ 9.
Polyrhachis pressa Mayr, Reise Novara. Zool., 2. Formicid., 1865, p. 39, ^

9 , pi. 1, fig. 5.

Type-locality: Batavia, Java (Novara Expedition).
A single worker from Kuehing (John Hewitt).

215. Polyrhachis (Myrmhopla) abdominalis Smith.

Polyrhachis abdominalis Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 63, ^ .
Polyrhachis phyllophilus Smith, Journ. Proc. Linn. soc. London. Zool. Suppl.,

1860, 4, p. 69, y .
Polyrhachis phyllophila Mayr, Tijdschr. ent., 1867, 10, p. 47, ^ .

Type-locality: Burmah.

Sarawak (Doria and Beccari); Kapouas Basin (Chaper).
Two workers from British North Borneo (E. B. Kershaw) and one
from Serambu Mt., Sarawak (H. W. Smith).

216. Polyrhachis (Myrmhopla) rubiginosa (Le Guillou).

Formica rubiginosa Le Guillou, Ann. Soc. ent. B'rance, 1841, 10, p. 316, ^ .
Polyrhachis rubiginosa Emery in Dalla Torre, Cat. Hymenop., 1893, 7, p. 268.

Type-locality: Borneo (Voyage of the "Astrolabe" and "Zelee").

*217. Polyrhachis (Myrmhopla) rugifrons Smith.

Polyrhachis rugifrons Smith, Journ. Proc. Linn. soc. London. Zool., 1861, 5,
p. 70, 9 ; Mayr, Tijdschr. ent., 1867, 10, p. 43, U .

Type-locality: Makassar, Celebes (A. R. Wallace).

A single worker from Serambu Mt., Sarawak (H. ^Y. Smith).

wheeler: the ants of borneo. 131

*218. PoLYRHACHis (Myrmhopla) furcata Smith.

PolyrhacMs f meatus Smith, Cat. Hymenop. Brit. mus. 1858, 6, p. 64, ^ , pi. 4,

fig. 20.
PolyrhacMs furcata Forel, Journ. Asiat. soc. Bengal, 1886, 55, p. 241, S ;

Emery, Ann. Mus. civ. Geneva, 1889, ser. 2, 7, p. 518, S 9 .

Type-locality: Burmah.

A single dealated female from the Rambungan River, Sarawak
(H. W. Smith). As Emery has observed, the petiolar spines of the
female are not hooked as in the worker, but merely curved.

219. Polyrhachis (Myrmhopla) chalybea Smith,

PolyrhacMs chahjbea Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 61, ^ .

Type-locality: Singapore (A. R. Wallace).

Sarawak (Doria and Beccari); Kapouas Basin (Chaper).

A single worker from Kuching (John Hewitt).

220. Polyrhachis (Myrmhopla) argentea Mayr.

Polyrhachis argenteus Mayr, Verh. Zool. hot. gesellsch. Wien, 1862, 12, p. 682,

^.
Polyrhachis argentea Mayr, Reise Novara. Zool., 2. Formicid., 1865, p. 40,

S , pi. 2, fig. 7.
Polyrhachis acastaForel, Journ. Asiat. soc. Bengal, 1886, 55, p. 241, ^ ; Dalla

Torre, Cat. Hymenop., 1893, 7, p. 257.

Type-locality: Manila, Philippine Islands (Novara Expedition).
Sarawak (Bedot and Pictet).

221. Polyrhachis (Myrmhopla) bicolor Smith.

Polyrhachis bicolor Smith, Cat. Hymenop. Brit. mus. 1858, 6, p. 65, 9 ; Mayr,
Verh. Zool. bot. gesellsch. Wien, 1862, 12, p. 681, 9 , pi. 19, fig. 5.

Type-locality: Burmah.

Sarawak (Doria and Beccari; Bedot and Pictet). Tandjong, S. E.
Borneo (Fritz Suck).

A few workers, females, and males from Kuching (John Hewitt).

132 bulletin: museum of compakative zoology.

221a. PoLYRHACHis (Myrmhopla) bicolor Smith var. aurinasis

Forel.

Polyrhachis bicolor var. aurinasis Forel, Mitth. Naturh. mus. Hamburg, 1901,
18, p. 77, y .

Type-locality: Sarawak, Borneo (Haviland).

222. Polyrhachis (Myrmhopla) cephalotes Emery.

Polyrhachis cephalotes Emery, Rev. Suisse zool., 1893, 1, p. 199, ^ , pi. 8, fig. 6;
Forel, Rev. Suisse zool., 1913, 21, p. 665, ^ 9 cf .

Type-locality: Deli, Sumatra (Bedot and Pictet).
Balik Papan, Borneo (Kampmeinert).

223. Polyrhachis (Myrmhopla) dives Smith.

Polyrhachis dives Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 64, ^ .
Polyrhachis acantha Smith, ibid, Suppl., 1860, 4, p. 98, ^ , pi. 1, fig. 16.

Type-locality: Singapore (A. R. Wallace).
Borneo (Leyden Museum).

*224. Polyrhachis (Myrmhopla) tibialis Smith.
Polyrhachis tibialis Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 63, ^ .

Type-locality: Burmah.

A single worker from Kuching (John Hewitt).

225. Polyrhachis (Myrmhopla) diotima Forel.
Polyrhachis diotima Forel, Rev. Suisse zool., 1911, 19, p. 60, ^ .

Type-locality: Hayvep, Borneo (Winkler).

226. Polyrhachis (Myrmhopla) aspasia Forel.
Polyrhachis aspasia Forel, Rev. Suisse zool., 1911, 19, p. 59, ^ .

Type-locality: Sarawak, Borneo (Haviland).

wheeler: the ants of Borneo. 133

*227. POLYRHACHIS (MyRMHOPLa) DAPHNE, Sp. nov.

Worker. Length: 5.5-6 mm.

Head a little longer than broad, a little broader at the eyes than at the
anterior corners, semicircularly rounded behind, with rather straight sides and
cheeks ; in profile very convex in the region of the frontal carinae, with feebly
convex gular surface. Eyes large and convex, their anterior orbits at the
median transverse diameter of the head. Palpi long and slender. Mandibles
moderately convex, with four stout, subequal teeth. Clj^peus semicircular,
convex, ecarinate; its anterior border very feebly sinuate on each side of the
middle which does not project as a lobe. Clypeal fossa very pronounced.
Frontal area indistinct, triangular. Frontal carinae approximated, sinuate,
erect, not further apart behind than in front. Antennae long, their scapes
extending fully half their length beyond the occipital border of the head.
Thorax subcylindrical, narrower than the head, a little broader through the
posterior part of the pronotum than through the meso- and epinotum which
are of equal width; pro- and mesoepinotal sutures very distinct and strongly
impressed. Pronotum, excluding the neck, a little broader than long, rounded
on the sides, with two stout, straight, acute spines which are twice as long as
the width of their bases and directed forward and outward. Mesonotum as
long as broad, its anterior outline semicircular, its lateral and posterior borders
straight, the sides not marginate. Epinotum compressed above, the base
horizontal and slightly concave, oblong, strongly marginate on the sides, the
marginations continued into the spines which are stout, straight, acute, twice
as long as those on the pronotum and as long as the base of the epinotum,
laterally compressed, and directed backward and very slightly upward. The
epinotal declivity is abrupt, slightly concave in profile, not marginate on the
sides, with large, prominent stigmata. Petiole subcuboidal with a denticle
on each side below near the anterior border and above with two stout, com-
pressed, curved spines, a little longer than those on the epinotum and directed
outward and backward. The distance between their tips is less than half the
diameter of the first gasti'ic segment. Gaster large, broadly elliptical, less
than half of it formed by the first segment, sUghtly flattened dorsoventrally,
convex but not truncated in front. Legs long; tibiae cylindrical, slightly
constricted at the base, without bristles on their flexor surfaces.

Mandibles and clypeus smooth and shining, the former minutely and sparsely
punctate; head shining above, subopaque on the sides; shagreened, the vertex
and sides of the front sparsely punctate. Thorax and petiole opaque; sides
of the former densely and evenly punctate, the dorsum very finely punctate-
rugulose, the rugules on the mesonotum obscurely concentric, the base of the
epinotum, especially behind, smooth and shining. Petiole indistinctly
punctate, slightly shining between the insertions of the spines. Gaster shin-
ing, very finely and superficially shagreened or aciculate, with small, sparse
punctures; legs more opaque and more coarsely shagreened.

134 bulletin: museum of comparative zoology.

Hairs yellow, erect; present only on the venter, tip of gaster, mandibles,
and anterior border of the clypeus. Pubescence yellow, very short, sparse,
and appressed, distinct on the sides of the head, on the legs and gaster.

Brownish yellow; mandibles reddish, with dark brown teeth; scapes,
except their tips, a median line on the clypeus, the upper surface of the head
and sides of the front, a large spot on the pronotum, the mesonotum, epinotum,
petiole, and basal half of first gastric segment, brown; epinotal and petiolar
spines, tarsi, and constricted bases of the tibiae somewhat darker.

Described from two specimens taken by Prof. Harrison W. Smith
on Matang Mt., near Kuching. Typc.— M. C. Z. 9,075.

I can find no description of this beautiful and striking species.

*228. PoLYRHACHis (Myrmhopla) personata, sp. nov.

Worker. Length nearly 5 mm.

Head oval, a little longer than broad, slightly narrowed in front and behind,
with rounded sides, convex above in the region of the frontal carinae, the lower
surface in profile convex in the middle, owing to a peculiar condition of the
occipital margin, which descends on each side of the narrowed gula as a promi-
nent ridge or fold and extends all the way to the anterior corner of the head,
with a blunt angle about half way between the neck and the mandible. The
ant therefore has the appearance of wearing a mask, the border of which stands
off on the sides and below from the neck. Eyes moderately large and convex,
broadly elliptical, with impressed orbits, at the middle of the sides of the head.
Mandibles feebly convex, 5-toothed. Clypeus moderately convex, subcarinate
behind, its border forming a short lobe, straight in the middle and rounded on
each side. Frontal area obsolete. Frontal carinae approximated, strongly
sinuate and bluntly angular in the middle, a little further apart behind than
in front. Antennae long, the scapes extending about -J their length beyond
the occipital border. Thorax subcylindrical, narrower than the head, about
twice as long as broad, slightly broader through the pronotum than elsewhere;
pronotum, excluding the neck, as long as broad, rounded and rather convex
above, anteriorly with a pair of straight, slender, acute spines, directed out-
ward, upward, and slightly forward. Promesonotal suture distinct and slightly
impressed; mesoepinotal suture obsolete. Mesonotum broader than long,
transversely rounded, its dorsal outline straight in profile. Epinotum short,
the base much shorter than the straight, abrupt declivity, armed with two
slender, acute spines about half again as long as those of the pronotum, di-
rected outward, backward, and upward, their tips very feebly recurved.
Petiole stout, with convex anterior and posterior surfaces, bearing above a
pair of long, stout, curved, acute spines. These are much longer and stouter
than those on the epinotum and form a semicircle, clasping the base of the
gaster. The median dorsal border of the petiole between their bases bears
two small acute teeth. Gaster subglobose, as broad as long, slightly flattened,

wheeler: the ants of borneo. 135

the first segment forming about f of its surface, truncated in front. Legs
moderately long; tibiae cylindrical, constricted at the base, without bristles
on their flexor surfaces.

Gaster and gula smooth and shining; remainder of body and the appendages
subopaque. Mandibles very finely striated and sparsely punctate; head,
thorax, and petiole denselS' punctate-rugulose, the rugules coarser and reticu-
late on the dorsal surface of the head; the spines of the epinotum and petiole
very finely shagreened, like the antennal scapes and legs.

Hairs whitish, erect; present only on the venter, tip of gaster, clypeal
border, and mandibles. Pubescence pale, very fine and dilute, visible only
on the tibiae, antennal funiculi, and sides of the gaster.

Black; palpi and spurs of the tibiae red.

Described from a single specimen taken by Mr. G. E. Bryant on
Mt. Matang, West Sarawak and sent me by Mr. Horace Donisthorpe.

This species belongs to the dives group but is very distinct in the
shape of the head, thorax, and petiole.

*229. PoLYRHACHis (Myrmhopla) atrovirens Emery.

Pohjrhachis alrorirens Emery, Ann. Mus. civ. Geneva, 1900, ser. 2, 20, p. 718,
U , fig. 16a.

Type-locality: Bua Bua, Engana (E. Modigliani).

A single worker from Sadong, Sarawak (H. W. Smith).

*230. PoLYRHACHis (Myrmhopla) oedacantha, sp. nov.
Worker. Length 4 mm.

Related to P. Mppomanes Smith and atrovirens Emery. Head elliptical,
longer than broad, semicircularly rounded behind with feebly convex sides
and equally convex dorsal and gular surfaces. Eyes moderately convex,
their anterior orbits at the median transverse diameter of the head. Mandibles
with rather rounded external borders, the apical borders with 5 subequal teeth.
Clypeus carinate at the base, its anterior border broadly rounded and some-
what crenulate. Frontal area distinct, triangular. Frontal carinae approxi-
mated, sinuous, a little further apart behind than in front ;^ frontal groove
absent. Antennae long and slender, their scapes extending fully -g- their
length beyond the posterior border of the head. Thorax from above elongate
trapezoidal, broadest through the humeri, evenly, longitudinally, and trans-
versely rounded and convex above, with distinct prom?sonotal and meso-
epinotal sutures. Pronotum about U times as broad as long, its anterior
corners produced as flattened, triangular and slightly upturned teeth. Meso-
notum of the same length as the pronotum; epinotum very short, without
distinct base, but sloping abruptly from the mesoepinotal suture which lies
just in front of the spines. These are long, stout, and acute, as long as the
declivity, widely separated and curved at their bases, with slightly sinuous

136 bulletin: museum of comparative zoology.

tips, directed backward, upward, and outward. Petiole in profile with straight,
not angulate, anterior and slightly convex posterior surface, armed above with
a pair of stout spines, longer and stouter than those of the epinotum, constricted
at the base and distinctly swollen in the middle, curved and tapering, directed
outward and backward around the base of the gaster. The distance between
their tips is a little less than the greatest diameter of the gaster, which is sub-
globose, distinctly broader than long, with | of its surface formed by the first
segment. Legs moderately long, tibiae cylindrical, strongly constricted at
their bases, without bristles on their flexor surfaces.

Head, thorax, petiole, and legs subopaque; mandibles very finely striated
and sparsely punctate; head, thorax, and petiole densely and* uniformly
punctate, thoracic dorsum also with scattered, shallower punctures; epinotal
and petiolar spines longitudinally rugulose. Gaster shining, more finely and
more superficially punctate than the head and thorax. Legs finely and sharply
shagreened.

Hairs and pubescence whitish, the former only on the venter, tip of gaster,
clypeus, and mandibles, the latter very fine, distinct only on the gaster, which
has a slightly pruinose appearance.

Black; head, thorax, and petiole with indistinct purplish metallic reflections,
more greenish on the occiput. Palpi, tibiae, femora, and middle and hind
coxae red, tips of femora and constricted bases of the tibiae more or less
infuscated.

Described from a single specimen taken by Mr. John Hewitt at
Kuching.

This species, though closely related to P. hippomanes, paromalus,
mucronata, and atrovirens, is easily distinguished by the robust, some-
what fusiform epinotal spines. It is perhaps merely a subspecies of
hippovianes, although the petiole does not agree with Smith's figure
of the type from Celebes. It is certainly very different from the subsp.
ceylonensis Emery, represented in my collection by a couple of speci-
mens received from Professor Forel.

23L PoLYRHACHis (Cyrtomyrma) rastellata (Latreille).

Formica rastellata Latreille, Hist. nat. fourmis, 1802, p. 130, 9 .
Polyrhachis rastellata Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 59.

Type-locality: East Indies (Riche).
Sarawak (Doria and Beccari).

wheeler: the ants of borneo. 137

232. PoLYRHACHis CCyrtomyrma) laevissima Smith.

Polyrhachis laevissimus Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 64, y ,

pi. 4, fig. 42.
Polyrhachis globularia Mayr, Tijdschr. ent., 1867, 10, p. 41, y .
Polyrhachis laevissima Mayr, Verb. Zool. bot. gesellsch. Wien, 1878, 28, p. 651,

9.
Polyrhachis levissima Dalla Torre, Cat. Hymenop., 1893, 7, p. 264.

Type-locality: Burmah.
Kapouas Basin, Borneo (Chaper).

233. Polyrhachis (incertae sedis) castaneiventris Smith.

Polyrhachis castaneiventris Smith, Cat. Hymenop. Brit, mus., 1858, 6, p. 67, 9 .
Type-locality: Sarawak, Borneo (A. R. Wallace).

234. Polyrhachis (incertae sedis) nitida Smith.

Polyrhachis nitidus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 61, 9.
Polyrhachis nitida Dalla Torre, Cat. Hymenop. 1893, 7, p. 266.

Type-locality: Sarawak, Borneo (A. R. Wallace).

235. Polyrhachis (incertae sedis) ruficornis Smith.

Polyrhachis ruficornis Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 60, 9 .

Type-locality: Sarawak, Borneo (A. R. Wallace).
This is probably the female of F. bicolor Smith.

236. Polyrhachis (incertae sedis) constructor Smith.

Polyrhachis constructor Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 60, 9 {nee P. constructor Smith, Cat. Hymenop. Brit, mus., 1858, 6,
p. 68).

Type-locality: Sarawak, Borneo (A. R. Wallace).

138 bulletin: museum of comparative zoology.

237. Echinopla lineata Mayr.

Echinopla lineata Mayr, Verb. Zool. bot. gesellsch. Wien, 1862, 12, p. 689, ^

Type-locality: Singapore.
Sarawak (Doria and Beccari).

238. Echinopla melanarctos Smith.

Echinopla melanarctos Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 79, y , pi. 25, fig. 29.

Type-locality: Singapore (A. R. Wallace).
Sarawak (Doria and Beccari; Haviland).

239. Echinopla pallipes Smith.
Echinopla pallipes Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 80, y .

Echinopla pallidipes Dalla Torre, Cat. Hymenop., 1893, 7, p. 272.

Type-locality: Sarawak, Borneo (A. R. Wallace).
A single worker from Kuching (John Hewitt).

240. Echinopla rugosa Ern. Andre.

Echinopla rugosa Ern. Andre, Mem. Soc. zool. France, 1892, 5, p. 47, y .
Type-locality: Kapouas Basin, Borneo (Chaper).

24L Echinopla striata Smith.
Echinopla striata Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,

p. 80, y .

Type-locality: Malacca (A. R. Wallace).

Sarawak (Doria and Beccari).

Three workers from Serambu Mt., Sarawak (H. W. Smith).

wheeler: the ants of borneo. 139

242. EcHiNOPLA sucKi Forel.

Echinopla siicki Forel, Mitth. Naturh. mus. Hamburg, 1901, 18, p. 75, y

Type-locality: Tandjong, N. E. Borneo (Fritz Suck).
Three workers from Kuching (John Hewitt).

243. Echinopla tritschleri Forel.

Echinopla tritschleri Forel, Mitth. Naturh. mus. Hamburg, 1901, 18, p. 74, ^ 9 .

Type-localit}/: Indrapura, Sumatra (Tritschler).
Tandjong, S. E. Borneo (Fritz Suck).

BoRNEAN Species Incertae Sedis.

244. Cerapachys oculatus Smith.

Cerapachys ocxilatus Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 74, c?.

Type-locality: Sarawak, Borneo (A. R. Wallace).

245. Ponera apicalis Smith.

Ponera apicalis Smith, Journ. Proc. Linn. soc. London. Zool. 1857, 2, p. 66, 9 .
Type-locality: Sarawak, Borneo (A. R. Wallace).

246. Ponera pompiloides Smith.

Ponera pompiloides Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2,
p. 69, cf.

Type-locality: Sarawak, Borneo {X. R. Wallace).

247. Ponera vidua Smith.

Ponera vidua Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. OS, cf .
Type-locality: Sarawak, Borneo (A. R. Wallace).

140 bulletin: museum of comparative zoology.

. 248. Atta penetrans Smith.

Atta -penetrans Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 77, 9
Aphaenogaster penetrans Dalla Torre, Cat. Hymenop., 1893, 7, p. 104.

Type-locality: Sarawak, Borneo (A. R. Wallace).

249. Atta cingulata Smith.

Atta cingulata Smith, Journ. Proc. Linn. soc. London. Zool., 1857, 2, p. 77, S
Aphaenogaster cingulata Dalla Torre, Cat. Hymenop., 1893, 7, p. 102.

Type-locality: Sarawak, Borneo (A. R. Wallace).

POSTSCRIPT.

A series of specimens, which Prof. C. F. Baker of Manila, P. I.
collected at Sandakan, Borneo and sent me too late to be included
in the foregoing revision, comprises several interesting and four new
species.

L Cerapachys antennatus Smith. {Ante p. 45).

Male. Length about 7 mm.

Long and slender. Head, including the eyes, as broad as long, the eyes
large, fully f as long as the sides of the head and situated a little in front of its
middle. Mandibles shaped as in the worker, their apical borders well-de-
veloped and finely denticulate. Clypeus indistinctly carinate behind, its
anterior border broadly rounded and entire. Antennae long and stout, scapes
somewhat longer than the three basal funicular joints together; first funicular
joint as long as broad, the second broader than long, the remaining joints
longer than broad, gradually increasing in length to the tip. Thorax through
the wing-insertions as broad as the head through the eyes, mesonotum feebly
convex, with pronounced Mayrian furrows. Epinotum in profile angular,
with subequal base and declivity, the latter flat, seen from behind subcircular,
strongly marginate both above and on the sides. Petiole subcylindrical, If
times as long as broad, sharply truncated and marginate in front. Post-
petiole broader than the petiole, broader behind than in front, also If times
as long as broad, with evenly convex dorsal, lateral, and ventral surfaces.

wheeler: the ants of Borneo. 141

Gaster very long and narrow, broadest in the middle, the basal half formed
by the first segment. Genital appendages sm^l and retracted. Legs rather
long and slender.

Shining; mandibles very sparsely, coarsely, and irregularly punctate.
Clypeus feebly striolate. Head finely and rather regularly longitudinally
rugose on the front, irregularly punctate-rugulose behind. Thorax coarsely
and transversely foveolate along the sutures and Mayrian farrows and coarsely
and sparsely foveolate on the surfaces of the sclerites. Pronotum longitudi-
nally rugose. Base of epinotum coarsely and vermiculately rugose, the
declivity finely and densely punctate-rugulose. Petiole above with a few
strong, undulating longitudinal rugae, becoming more reticulate on the sides.
Postpetiole, gaster, and legs smooth, Avith minute, sparse, piligerous punctures.

Hairs as in the worker, and most abundant on the apical portion of the gaster.

Black; mandibles, genitalia, and tips of tarsi reddish. Wings grayish
hyaline, with large black pterostigma and dark brown veins.

Described from a single specimen, which must, I believe, represent
the hitherto unknown male of C. antcnnatus though the sculpture of
the head, thorax, and petiole is very different from that of the worker.

*2. Myopopone castanea Smith 9 •
3. Platythyrea pusilla Emery 9 {Ante, p. 50).
*4. Bothroponera sandakana, sp. nov.

Female. Length about 14 mm.; wings 11 mm.

Allied to rufipes Jerdon and insularis Emery. Head, excluding the mandi-
bles, broader than long and considerably broader behind than in front, with
broadly excavated posterior border and nearly straight sides. Posterior
orbits of the moderately large and convex eyes at the median transverse
diameter of the head. Mandibles large and rather convex above, not flattened
as in rufipes, their apical bowlers finely and rather evenly serrate. Clypeus
short, strongly carinate, truncated anteriorly in the middle; the anterior border
entire, straight, and transverse. Antennae short and stout, scapes reaching
nearly to the middle of the occipital bsrder, all the funicular joints, except the
first and last, distinctly broader than long. Thorax through the wing-inser-
tions a little narrower than the head. Epinotum sharply angular in profile,
the base feebly convex and distinctly shorter than the concave, sloping de-
clivity, which has a marked, crenulate border, both above and on the sides.
Petiole higher than thick, as thick above as below, truncated in front and be-
hind, convex above and on the sides, the posterior surface slightly concave
but with its superior border merely sub marginate and not denticulate. Gaster
and legs of the usual form.

142 bulletin: museum of comparative zoology.

Mandibles shining, very coarsely and sparsely punctate, with a few coarse
rugae near the base. Remainder of body subopaque, densely punctate; head,
thorax, and petiole also covered with sparse but rather regular foveolae, which
are more pronounced on the posterior portion of the head, sides of thorax,
and on the petiole. Epinotal declivity rather shining, very minutely punctate.
Postpetiole and first gastric segment longitudinally but much less sharply
costate than in rufipes, the spaces between the costae with shallow foveae
having sharper anterior margins. These foveae become much more distinct
on the sides of the segments. Legs with sparse piligerous punctures.

Hairs and pubescence golden, abundant and rather long, the pubescence
appressed on the head, gaster, and appendages, more oblique on the thoracic
dorsum, nearly absent on the pleurae. Hairs longest on the abdomen, espe-
cially on the apical segments.

Dark brown; mandibles blackish; legs, including the coxae, posterior
borders of gastric segments, wing-insertions, epinotum and petiole reddish.
Wings rather heavily infuscated, with blackish stigma and brown veins.

A single specimen. I describe this large ant with some reluctance.
It is certainly not a form of rufipes or of insularis, judging from
Emery's very brief description of the worker of the latter species.
Professor Baker sent me from Singapore a female specimen which
evidently represents a variety of insulcn-is, and I possess a worker
from Ceylon labeled " insnlaris" by Forel. Though these specimens
have the mandibles much like those of sandakana in being shining,
convex, and without the fine, longitudinal striae of rvfipes, the petiole
has a sharp denticulate postero-superior border. According to Emery,
the petiole is also costate in the typical insularis and this is the case
in my Ceylonese specimen.

5. Bothroponera tridentata (Smith) ^ cf (Ante, p. 55).

Male. Length nearly 10 mm.

Head through the very large, convex, and reniform eyes broader than long,
flat above, with evenly rounded posterior border and very short cheeks, the
eyes occupying nearly the whole of the sides of the head. Ocelli large and
prominent. Mandibles small, flat, edentate. Palpi very long, the labial pair
3-, the maxillary pair 5-jointed, the three terminal joints of the latter long and
attenuated. Clypeus feeblj^ convex, ecarinate, with straight, entire anterior
border. Antennae very long and slender, filiform; scape short, only twice as
long as broad, the first funicular joint as broad as long, the remaining joints
cyhndrical, gradually diminishing in length towards the tip. Thorax somewhat
broader through the wing-insertions than the head. Pronotum transverse,
truncated in front; mesonotum as long as broad, feebly convex above, without

wheeler: the ants of borneo. 143

Mayrian furrows; scutellum very convex; epinotum in profile strongly angu-
lar, both the base and declivity concave, the former shorter than the latter,
the two surfaces separated by a pronounced, subcircular margination or carina.
Petiole like that of the worker but with the three blunt teeth on the postero-
superior border much less prominent, the ventral surface with two strong
teeth, the anterior directed downward and forward, the posterior somewhat
longer and more slender, directed downward and backward. Postpetiole
broader than long, evenly and convexly rounded in front, strongly marked
off from the broader and very short gaster, the last segment of which terminates
in a long, stout, downwardly curved spine. GenitaUa deeply retracted. Legs
slender.

Subopaque; postpetiole, gaster, and legs shining. Head very finely punc-
tate; thorax densely punctate, the pro- and mesonotum and mesopleurae
also with indistinct, scattered foveolae. Scutellum and epinotum irregularly
rugose, the declivity of the latter more finely, so that its surface is somewhat
shining. Petiole coarsely reticulate-rugose, its truncated posterior surface
finely rugulose. Postpetiole, gaster, and legs smooth, with fine piligerous
punctures.

Hairs and pubescence as in the worker, but shorter.

Head, postpetiole, and gaster castaneous; thorax and petiole black, neck
and discal portion of mesonotum red; antennae, palpi, and legs, including
the coxae, yellow, the tibiae streaked with fuscous or black. Wings clear yel-
lowish hyaline; pterostigma brown, veins yellow.

A single specimen, which I have described in detail, because very
few Bothroponera males have been seen.

*6. Pseudoponera amblyops (Emer}') 9 .
*7. Trapeziopelta breviloba, sp. nov.
Female. Length 5.8 mm.

Head, excluding the mandibles, a little broader than long, slightly broader
behind than in front, with feebly rounded sides and feebly and broadly excised
occipital border. Eyes rather convex, longer than their distance from the
anterior corners of the head. Ocelli small and close together. Mandibles
distinctly shorter than the head, slender and terete at the base, dilated and
flattened apically, their inner borders with three separated teeth; /the most
basal merely a low, rounded convexity, the others stronger and decidedly
acute. Lobe of clypeus very short, rectangular, more than twice as broad as
long, its anterior border slightly concave. Oral border of clypeus with a
median, slender, truncated tooth. Frontal groove strongly impressed. An-
tenna] scapes not reaching to the posterior border of the head; funiculi with
distinctly 4-jointed club; joints 2-7 slightly broader than long; joints 8-10
onger than broad; terminal joint nearly as long as the three remaining joints

144 bulletin: museum of comparative zoology.

of the club together. Thorax 2| times as long as broad, parallel-sided, dis-
tinctly narrower than the head, flattened above; pronotum, excluding the
neck, about as long as the mesonotum; base and declivity of epinotum forming
nearly a right angle in profile, the declivity slightly concave. Petiole, post-
petiole, and gaster together but little longer than the thorax, the petiole trun-
cated anteriorly and posteriorly, higher than long, from above broader than
long and broader behind than in front, with rounded dorsal and lateral sur-
faces, its ventral surface in front with a blunt, compressed tooth. Postpetiole
rather strongly constricted behind, its anteroventral surface with an acute,
downwardly directed tooth. Sting long and compressed.

Smooth and shining; head, thorax, and abdomen with small, sparse, incon-
spicuous, piligerous punctures.

Hairs yellow, bristly, pointed, of uneven length; sparser on the body and
legs, suberect on the former, oblique on the latter; more abundant but short
on the antennae.

Deep red; antennae, legs, and mandibles slightly yellowish red. Wings
uniformly brown, with dark brown veins and conspicuous black pterostigma.

Described from a single specimen.

This does not seem to be the female of any of the described species,
nearly all of which are known only from worker specimens. It is
evidently moet closely related to the Papuan T. laevigata Emery, the
female of which is still to be discovered.

8. DoRYLUS LAEViGATUS (Smith) cf. {Ante, p, 61).
9. Tetraponera attenuata Smith ^ 9 d^. (Ante, p. 65).

10. Tetraponera difficilis Emery (?) 9 . {Ante, p. 65).

11. Tetraponera pilosa (Smith) ^ 9. {Ante, p. 65).

12. Myrmicaria arachnoides (Smith) subsp. melanogaster Emery

9 . {Ante, p. 71).

13. Vollenhovia rufiventris Forel 9 . {Ante, p. 79).

*14. Vollenhovia oblonga Smith var. (?) 9 .

15. Pristomyrmex trachylissus (Smith) 9 . {Ante, p. 86).

*16a. Paratopula, gen. nov. ceylonica (Emery) var. sumatrensis

(Forel) 9.

I have received all three phases of this ant from the Philippines
(F. X. Williams). The typical form of the species, originally described

wheeler: the ants of borneo. 145

as Atopomyrmex ceylonicus, has recently been placed by Forel in the
genus Leptothorax. As I cannot agree with this allocation and
as the species cannot be left in Atopula, I have coined a new generic
name, Paratopula.

17. Meranoplus mucronatus Smith ^ . (Ante, p. 90).
18. Cataulacus granulatus (Latreille) 9 . {Ante, p. 92).

19. Cataulacus hispidulus Smith ^ . {Ante, p. 93).-

20. Cataulacus latissimus Emery ^ . {Ante, p. 94).

*21. Myrmoteras bakeri, sp. nov.

Female. Length nearly 4 mm.

Very similar to M. donisthorpei Wheeler, but differing in the following
characters: the five large teeth along the apical half of the mandibles are
distinctly longer and stouter, though alternating with shorter teeth as in
donisthorpei, whereas the denticles on the basal half are very small and almost
obsolete; the frontal carinae are somewhat further apart; the lateral lobes
of the head just behind the eyes are more acutely angular; the petiolar node
is much more compressed anteroposteriorly so that its upper border is trans-
verse and rather sharp; the middle and hind tibiae are less strongly swollen
in the middle and the sculpture and color are different. The head is very
smooth and shining like the remainder of the body, and the insect is honey-
yellow, with the mandibles, except their brownish teeth, the femora and tarsi
paler, whitish yellow. The middle portion of the first gastric segment and the
bases of the succeeding segments are brownish. The pilosity and wing-
venation are as in donisthorpei, but the erect hairs on the scapes, body and legs
are somewhat coarser and more bristly.

Male. Length 3 mm.

Head through the eyes broader than long, gradually contracted and rounded
behind, without the peculiar lobular eminences of the female, to the concave
and marginate occipital border. Eyes large, but placed far forward, so that
the cheeks are very short. Ocelli small. Mandibles very small, vestigial,
bluntly pointed, edentate. Clypeus, frontal carinae, and antennae much as
in the female, except that the antennae are 13-jointed. Thorax, gaster, and
legs as in the female; middle and hind tibiae less incrassated, petiolar node
thicker and much blunter above. Genital appendages small, exserted, super-
ficially like those of Prenolepis. Wings as in the female, with the same
peculiar venation.

146 bulletin: museum of comparative zoology.

Smooth and shining; thorax subopaque, finely punctate-rugulose.
Pilosity as in the female but somewhat sparser and finer.
Castaneous brown; pronotum and head paler and more reddish; palpi and
tarsi beyond the first joint, whitish.

Described from a single female and three males. More material may
show that this form is merely a pale race, or subspecies of donisthorpei.
The color of the female before me is certainly not due to immaturity.
Dr. F. X. Williams has recently sent me a male and female of another
species, M. iviUiamsi, sp. nov., from the Philippines, the fourth species
of this remarkable genus to come to light in the Malayan subregion.
The female is larger and more robust than donisthorpei and bakcri,
with the head and thorax rich reddish brown, opaque, and very finely
and densely punctate and the basal half of the swollen middle and
hind tibiae black, the apical half yellow. The male is black, with pale
terminal tarsal joints and the head and thorax are densely punctate-
rugose. The worker is known only of the type-species, M. binghami
Forel of Burma.

22. Camponotus (Myrmotarsus) mistura (Smith) y .

(A7ite, p. 109).

*23. Camponotus (Myrmotarsus) satan, sp. nov.

Female. Length: 18 mm.; wings 19 mm.

Head subtrapezoidal, broader than long without the mandibles, much
broader behind than in front, with straight sides and broadly excised posterior
border. Eyes moderately large and convex. Mandibles large, convex,
with 6 large, subequal teeth. Clypeus broader than long, rather flat, ecari-
nate; its anterior margin straight and transverse in the middle, with a small
tooth in each side and feebly concave lateral to each tooth. Frontal carinae
slightly diverging behind and not strongly curved . Antennal scapes decidedly
flattened, reaching to the posterior corners of the head. Thorax and legs as
in the other species of the subgenus. Petiole broad, cuneate in profile, its
anterior and posterior surfaces flat, its superior margin moderately sharp,
feebly and sinuately emarginate. Gaster considerably shorter than the
thorax. Tibiae and middle and hind metatarsi flattened as in other species
of the subgenus. Wings long.

Very smooth and shining except thei sides of the thorax and the top of the
head, which are opaque. Mandibles, clypeus, and cheeks sparsely punctate,
the punctures becoming finer and denser on the sides of the head; the opaque
dorsal portion sharply and finely coriaceous.

wheeler: the ants of Borneo. 147

Hairs black, rather short, coarse, moderately abundant, erect, absent on
the cheeks but covering the body and appendages, short on the scapes ; venter
and borders of gastric segments above, and lower surfaces of tarsal joints with
short, appressed fulvous hairs. Pubescence absent.

Deep black; terminal tarsal joints reddish. Wings brown with resin colored
veins and apterostigma bordered with darker brown.

A single specimen. This species is quite distinct in the shape of
the head, sculpture and color of the pilosity.

24. Camponotus (Myrmosphincta) camelinus (Smith) S .

(Ante, p. 111).

24a. Camponotus (Myrmosphincta) camelinus var. singularis

(Smith) S . (Ante, p. 112).

25. Polyrhachis bihamata (Drury) S 9 . (Ante, p. 121).

26. Polyrhachis (Myrma) vindex Smith S . (Ante, p. 125).

A single specimen, differing from the typical form described on
p. 125 in having the legs entirely black.

27. Polyrhachis (Myrmhopla) armata (Le Guillou) ^ . (Ante,

p. 129).

28. Polyrhachis (Myrmhopla) furcata Smith 9 . (Ante, -p. 131).

29. Polyrhachis (Myrmhopla) chalybea Smith S . (^Antc, p. 131).

30. Polyrhachis (Myrmhopla) cephalotes Emery ^ . (Ante,

p. 132).

31. Polyrhachis (Cyrtomyrma) rastellata (Latreille) S .

(Ante, p. 136).

A specimen with the legs entirely black.

32. Echinopla pallipes Smith ^ . (Aiitc, p. 138).

.r.r>

Bulletin of the Museum of Comparative Zoology
AT HARVARD COLLEGE.

Vol. LXIII. No. 4.

REPORTS ON THE SCIENTIFIC RESULTS OF THE EXPEDITION TO THE
TROPICAL PACIFIC IN CHARGE OF ALEXANDER AGASSIZ, ON THE U. S.
FISH COMMISSION STEAMER "ALBATROSS," FROM AUGUST, 1899, TO
MARCH, 1900, COMMANDER JEFFERSON F. MOSER, U. S. N., COMMANDING-

XXI.

;

THE BIRDS.

By Charles Haskins Townsend and Alexander Wetmore.

Published by Permission of H. M. Smith, U. S. Fish Commissioner].

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

August, 1919.

No. 4. — Reports on the scientific results of the Expedition to the Tropical
Pacific in charge of Alexander Agassiz, on the U. S. Fish Com-
mission Steamer Albatross, from August, 1899, to March, 1900,
Commander Jefferson F. Moser, U. S. N., Commanding.

XXI.

The Birds.

By Charles Haskins Townsend and Alexander Wetmore.

Part 1. — Introduction and Field Notes. By Charles Haskins

Townsend.

The voyage of the U. S. Fisheries Steamer Albatross during the
winter of 1899 and 1900, under the direction of Mr. Alexander Agassiz,
was undertaken for the purpose of studying the coral-reef districts of
the Tropical Pacific, making deep-sea explorations in the great ocean
basins, and conducting investigations of the fisheries among the islands.
A minor object of the expedition was the study of the fauna and flora
of oceanic islands, but it became necessary to restrict this part of the
program in order to investigate more thoroughly the structure and
formation of coral reefs and, islands.

The aciivities of the scientific staff took various directions at the
islands visited, ornithology receiving some attention. The collection
of birds brought back, ninety-three distinct forms and 406 specimens,
does not represent any systematic effort to secure birds. It is rather
the result of opportunities embraced from time to time, as the progress
of the expedition permitted. Our stops were generally brief, except
at points where coal or other supplies were renewed and some of these
were islands already well knowTi to ornithologists.

The preparation of bird skins being largely dependent upon the
personal efforts of the writer, there was naturally a limit to the number
which could be preserved. This usually meant night work, after the
ship's laboratory had been cleared of the day's deep-sea dredgings,
or the marine gatherings from the reefs. In the pleasanter work of
collecting, assistance was more often available. Occasionally it
became necessary to store birds in the ship's ice room for several days
until an opportunity could be found to preserve them properly.

Specimens were obtained on thirty-three different islands. It is to

152 bulletin: museum of comparative zoology.

be regretted that more time could not have been spent in exploring
certain well-forested islands, where unknown land birds may exist.
The atolls and reefs yielded little but the widely distributed water birds.

The time of year spent among these islands lying so near the equator,
that is our winter season, may explain the fact that no nests of resident
land birds were seen.

It is evident that observations on the habits of birds could seldom
be made during such hurried trips as we made ashore.

The expedition proceeded in turn through the Marquesas, Paumotu,
Society, Cook, Tonga, Fiji, Ellice, Gilbert, Marshall, Caroline, and
Ladrone Archipelagoes. These have been classified as Eastern
Polynesia, comprising the Marquesas, Paumotu, Society, and Cook
Groups; Central Polynesia, including the Samoan, Tonga, and Fiji
Groups; the Central Coral Islands, of the Ellice, Gilbert, and Marshall
Atolls, and Northwestern Polynesia, with the Caroline and Ladrone
Groups.

Although the collection of birds brought back by the Albatross is
far from being a representative one, new species were met with in
more than half of the groups visited, a fact indicating that there is still
much to be learned about the birds of Polynesia. Some of the larger
and better known islands have been but little explored ornithologically,
and many of the smaller ones not at all. It is probable that the Fiji
Archipelago with its 150 islands will eventually yield new birds.

Our knowledge of Polynesian birds does not extend much further
back than the time of the United States Exploring Expedition, 1838
to 1842, under Captain Charles Wilkes, when extensive collections
were made by Titian R. Peale, Zoologist of the Expedition.

A period of about a quarter of a century appears to have elapsed
before ornithological researches in this region were resumed. Since
then there have been many contributions to the ornithology of Poly-
nesia. More than 200 species of strictly land birds are now known to
the archipelagoes visited by the Albatross. As many of the genera
prevail for thousands of miles through the region, the presence of its
avifauna may be explained as the result of immigration. The dis-
tribution of a few species has been extended through the agency of
human beings.

The long-delayed appearance of this report is due to the fact that
the present writer left ^Yashington soon after the return of the Alba-
tross and never had an opportunity to study the collection of birds
brought back and placed in the U. S. N. M. He is deeply indebted to
Mr. Wetmore for the careful study he has made of the material.

TOWNSEND AND WETMORE : THE BIRDS. 153

The Albatross sailed from San Francisco on 23 August, 1899,
arriving at the Marquesas Islands 15 September, having occupied
twenty-seven sounding and dredging stations in depths varying from
687 to 3,088 fathoms on the way. Hydrographic work was carried
on not only during the passage from one archipelago to another, but
also at points among the islands composing the different groups.

The Marquesas Islands. — These are all high islands of volcanic
origin and without coral reefs. The group consists of nine principal
islands, with some outlying islets of small size.

The ornithology of the Marquesas, with the exception of Nuku-
hiva and Hiva-oa, is unknown. Most of the eleven land birds previ-
ously known came from the latter and its outlying island Tahuata or
St. Christina. All of the larger islands are forested and may yield
much of interest to naturalists. These are Motu Hiva, eight miles
in length by four in width and over 3,600 feet high; Tau Ata, nine
miles long by five wide and 3,200 feet high; Na-pu, about seven miles
in diameter with a height of 4,000 feet; Ua-huka, measuring seven
by five miles, and Eiao, six miles long and 2,000 feet in height.

Nukuhiva, also called Marchand, was the only island of the Mar-
quesas Group visited by the Albatross, the vessel lying at anchor in
the harbor of Tai-o-hae from 15-17 September. It is the largest of
the Marquesas, being fourteen miles in length and ten in width. The
island is mountainous, one peak rising to a height of 3,900 feet, with
others nearly as high, while most of it is very rugged. The heads of
some of the valleys are faced with steep cliffs, and there are many
cascades falling from considerable heights. There are fertile valleys,
little cultivated owing to a decreasing population, which now numbers
less than 1,000. Nukuhiva is well forested and has the fruits common
to Polynesian Islands.

Our stay was too short for anything more than desultory collecting
of birds in the vicinity of the anchorage. It is quite possible that
Nukuhiva has other birds than the species we obtained.

The land birds secured were Jungle cock (Gallus gallus), fruit
pigeon (Ptilopus dupetithouarsii) , swift {CoUocalia ocista), and warbler
(Conopoderas percernis), the last being a new species. All of these are
common. The water birds were Heteraditis incanus and Gygis
microrhyncha. The White noddy (Gygis viicrorhyncha) of which four
specimens were secured, was seen high up in the mountains, flying
across deep, forested ravines and occasionally observed resting on the
trees. It also nests in trees. This species is apparently restricted
to the Marquesas. Tropic birds (Phaethon rubricaudus) were also seen

154 bulletin: museum of comparative zoology.

in the mountains at Nukuhiva. It was not uncommon for them to
approach the vessel when we stopped for deep-sea sounding or dredg-
ing, and specimens shot from the deck could easily be recovered by
lowering a boat. The first tropic birds were observed 600 miles north-
east of the Marquesas.

Gallus has long been wild on the mountains of Nukuhiva. All the
specimens that I killed looked very much like the richly colored one
that was preserved. It was ver^*- interesting to hear the cocks crowing
in the high woods. I found that they could fly like pheasants, making
strong flights across wide ravines.

The large, thrush-like warbler of Nukuhiva {Conopodcras percerms)
is the most interesting bird of the island and by far the best songster
we heard in Eastern Polvnesia. It is not uncommon to hear several
of them singing loudly on a single tree. Herman Melville in "Typee"
that classic of the Marquesas, says, "Birds, bright and beautiful birds,
fly over the valley of Typee . . . but alas the spell of dumbness is
upon them all — there is not a single warbler in the valley." This
statement is altogether at variance with our experience. Everywhere
about the village of Tai-o-hae, only a few miles from the valley of
Typee, the woods rang with the melody of the warblers. This species
is the size of a mockingbird and its sulphur-yellow under parts render
it quite conspicuous.

The Paumotu Islands. — After leaving the Marquesas, the Alba-
tross proceeded through the Paumotu, Tuamotu, or Low Archipelago,
making brief stops at a dozen of the islands. There are seventy-eight
islands in this Group, the main body of which is more than a thousand
miles in length. With few exceptions they are atolls with enclosed
lagoons of considerable size. The atolls are usually well forested
with cocoanut trees, the principal species, and with low trees and
shrubs. They do not vary much in vegetation or general appearance.
The population is limited, the entire Archipelago having not more
than 4,000 inhabitants. The lagoons yield much valuable pearl shell,
but the export of copra is probabh' of greater importance.

Our first anchorage was at Rangiroa, where the Albatross re-
mained at anchor from 21-24 September. Rangiroa, also called
Rahiroa, Vliegen, Deans, and Nairsa, is an atoll with a great lagoon
more than forty miles in length. The land birds obtained were the
Blue lory {Choriphihis pcruxianus) and a warbler {Conopodcras aiijpha
nesiarcha), the latter described as new (p. 210).

The small Blue lory {Choriphihis peruriaruis) obtained at Rangiroa
and later at Bora Bora and Aitutaki, is a fairly common species and is

TOWNSEND AND WETMORE: THE BIRDS. 155

often kept as a pet by the natives, a custom which has aided its
distribution. One which I secured at Rangiroa became very tame,
and Hved on board the Albatross for several weeks, when it was
drowned by faUing into a pail of water. Its greatest delight was to
walk back and forth on a light rope stretched across my stateroom.
The water birds obtained were Hefcractitw incanm, Megahpterus
mclanogenys, Procclslcrna ceridea, and Sterna hinata.

Makatea, (Metia, Aurora, Recreation), where the Albatross
made a brief stop on 26 September, was revisited for a few hours on

6 October. It is one of the few elevated islands in the Paumotus,
having a height of over 200 feet, with perpendicular cliffs along the
north shore. It has a length of five miles and is densely wooded,
with many large trees. The higher parts are difficult to penetrate
with the thickly tangled \-egetation and massed pinnacles of sharply
weathered limestone. Our stay was too brief to permit of collecting
more than a few yards distant from the shore. The land birds found
here were fruit pigeon {Ptilopus coralensw), large pigeon (Globicera
aurorae), and warbler (Conopodcras atypha erema), the last being new.
All are common. The water birds were Pluvialis dominicus fulvus,
Phaeopus tahitiensis, and Anous stoUdus pileatus. This was our first
meeting with the large pigeon Globicera. It is rather common and
was observed later at Tahiti. This bird is called Ru-pee by the natives.
The dense forests of Makatea doubtless harbor species of land birds
which could only maintain themselves with difficulty in the scattered
cocoanut groves of the atolls.

NiAU, also called Creig and Faau, is an atoll about four miles in
diameter and is well wooded. It is somewhat higher than the neigh-
boring atolls, having an elevation of twenty-six feet. Our stop on

7 October of two hours afforded little time for collecting. The only
bird obtained was the fruit pigeon {Ptilopus coralensis) which is
abundant.

Apataki, sometimes called Hagemeister, is a partly wooded atoll
eighteen miles in diameter. Anchoring on the evening of 7 October
and sailing the following morning, only one bird was secured, a warbler
{Conopoderas atypha agassizi), which, however, proved to be unde-
scribed.

TiKEi, or Romanzoff, is a small, well-wooded island about three
miles in diameter and with no central lagoon. The Albatross
anchored at Tikei for a few hours on 8 October. The only birds
collected were Conopodcras atypha atypha and Pluvialis dominicus
fulvus. I found here the egg of a White-crowned black noddy balanced

156 builetin: museum of comparative zoology.

precariously upon the horizontal limb of a low Pandanus, the parent
bird covering it until I was almost near enough to seize it.

Fakaeava, (Fakarawa, Wittgenstein), is one of the more important
atolls, being about thirty miles in length and having tliree entrances
to the great lagoon available for vessels. It is rather irregularly
wooded. Although the ship remained from 10-14 October and
numerous birds were collected, there were only two species, Ptilopus
coralensis and Co^wj^dcras atypha atypha (new).

Although at work among the islands for several days after leaving
Fakarava, no anchorages were made until Makemo was reached on
19 October, where the vessel remained until the 25th. The only
land birds found were Ptilopvs coralensis and Conopoderas atypha
crypta (new). The water birds collected were Fregata arid ariel,
Plvvialis dominions fvhns, Phacopus iahitiensis, Heteractitis incamis,
Anovs stolidns pilcahis, Mcgaloptcr%is mclanogenys and Thalasseus
hergii rcdrirostris.

The Crested tern (Thalasseus hergii rectirostris) is rather common
among the Paumotu Islands. At Makemo we found it frequently
perching on stakes or piles along the shore of the lagoon near the
village. The other sea birds found here are common throughout the
Paumotus. Makemo, also called Philip Island, is a large atoll about
forty miles in length, the northern part of which is well wooded.

Tekokoto, (Tekareka, Doubtful Island), visited 26 October, is
nothing more than a tiny atoll a mile or so in diameter with a shallow
lagoon. It is only a few feet high, a part of it being covered with
bushes. These were heavily loaded with frigate birds and boobies,
great numbers of them fl.ying over the whale-boat as we searched for a
landing. The surf proved altogether too boisterous for safety and
the attempt was abandoned. We found frigate birds more numerous
at Tekokoto than at any other island \dsited during the voyage. The
natives of the Paumotus often keep tame frigate birds on perches near
their houses. The birds are reared in captivity and are used after
the manner of homing pigeons to carry messages among the islands.

It appears that the birds return promptly when liberated from quite
distant islands. They are distributed by being put aboard small
vessels trading among the islands. The birds are liberated whenever
there is news to be carried, returning to their perches sometimes in an
hour or less, from islands just below the horizon and out of sight of the
home base. Generally they are in no great hurry. As the food of a
frigate bird may be picked up almost anywhere at sea, there is no
means of ascertaining how much time the bird loses in feeding en route.

TOWNSEND AND WETMORE: THE BIRDS. 157

It may also linger to enjoy its liberty with other frigate birds. At
home it is usually tethered to its perch.

Mr. Louis Becke says they were used as letter carriers on the Samoan
Islands when he was there in 1882, carrying messages between islands
sixty or eighty miles apart. When he lived on Nanomaga in the
Ellice Islands, he exchanged two tame frigate birds with a trader
living on Nuitao, sixty miles distant, for a tame pair reared on that
island. The four birds at liberty frequently passed from one island
to the other on their own account, all going together on visits to each
other's homes, where they were fed by the natives on their old perches.
Mr. Becke's pair usually returned to him within from twenty-four to
thirty-six hours. He tested the speed of the "frigate" by sending
one of his birds by vessel to Nuitao where it was liberated with a
message at half-past four in the afternoon. Before six o'clock of the
same day the bird was back on its own perch at Nanomaga, accom-
panied by two of the Nuitao birds, which not being at their perch on
that island when it was liberated, it had evidently picked up on its
way home.

The tame frigate bird returns regularly to its home perch at night.
The use of the frigate bird as a carrier is referred to by the Rev. Dr.
George Turner in Samoa a Hundred Years Ago, page 282.

The Albatross did not anchor at Akiaki, but I made a hasty
landing 30 October, obtaining specimens of a warbler {Conopodcras
ati/pha rava), new. This island, also called Les Lanciers and Thrum
Cap, is less than a mile in diameter. It has no lagoon, is wooded and
is higher than the atolls.

On the 31 October the vessel reached Pinaki or Whitsunday Atoll,
but no anchorage was found and my own boat was the only one that
suceeded in making a landing through the dangerous waves that beat
upon the reefs. This wonderful circular atoll, which has often been
figured in works on geography, is ^a mile and a half in diameter, and is
well forested, especially with cocoanut and Pandanus. There is a
single shallow entrance to the enclosed shallow lagoon. The only
birds obtained were a warbler {Conopoderas atypha rava) and a single
sandpiper {Aechnorhynchus parvirostris) , one other being seen. The
latter species was not observed elsewhere and the only other specimens
known are those obtained by Peale on islands of the Paumotu Group.
Landings at Pinaki are difficult and the atoll is uninhabited. This
was the last island of the Paumotus on which birds were collected.

Although I made a landing on Here-here-tue 3 November, no land
birds were seen.

158 bulletin: museum of comparative zoology.

The Society Islands. — This group consists of ten principal islands
and several small islets. They are high, rugged, and forested volcanic
islands, usually with extensive coral reefs surrounding them. The
barrier reefs often enclose large lagoons.

The first island reached was Mehetia (Maitea, Osnaburg) where
I landed for half an hour while the ship lay to, but found no land birds.
Mehetia may be worth a careful examination by the ornithologist, as
it is seven miles in diameter, 1,400 feet high, and is well forested.
It is uninhabited. A specimen of Sula hxicogastra plotiis was obtained
at this island.

Tahiti, formerly called Otaheiti, is the largest and most important
of the Society Islands. It is in fact the largest in Eastern Polynesia,
excepting of course the Hawaiian Islands which constitute an entirely
separate group. Its greatest diameter is thirty-three miles and the
highest peak has an elevation of 7,321 feet. The island is heavily
forested, well watered, and has many fertile valleys which yield an
abundance of the plant products of the tropics. There are numerous
streams, many of which have at their heads waterfalls of great height.
The population of Tahiti probably exceeds 10,000.

The Albatross visited Tahiti twice, 27 September to 5 October,
and again 6-15 November. The accumulated marine collections of
the vessel were occupying so much space on board that it became
necessary to pack and ship them from this point. This heavy task
naturally interfered with the collecting of birds. Only six species of
land birds were secured, fruit pigeon {Ptilopus purpuratus), kingfisher
{Todirhamphus veneratus), swifts (CoUacalia ocista, and CoUocalia
thcspcsia), weaver birds {Lonchiira caManeothorax, and Acgintha
temporalis), the last two being introduced. The native name for
Todirhamphus is Ru-ru. No attempt was made to get water birds.
Tahiti is poor in land birds, there being but few species besides those
we obtained. One of these, the large pigeon {Globiccra aurorae),
we met with at Makatea in the Paumotus. The Polynesian duck
{Anas supcrciliosa) , was seen in the crater lake of Vaihiria, where a
hawk, said to have been introduced, was also seen. The frigate bird,
tropic bird, and White noddy were found high up among the moun-
tains. The kingfisher, Todirhamphus, is common in the heavily
wooded sections. So far as our examinations show, it feeds entirely
on insects, although frequenting the vicinity of streams inhabited
by small fishes.

The next anchorage after leaving Tahiti was at Bora Bora (Bola
Bola) 17-18 November, where three species of birds were secured,

TOWNSEND AND WETMORE : THE BIRDS. 159

Blue lory {Choriphilus peruvianus), kingfisher {Todirhamphus tutus),
and the introduced weaver bird (Lonckura castaneothorax) . A whole
afternoon of diligent exploration failed to reveal any other species.
The Blue lory, Choriphilus, is called Ve-ne, and the kingfisher, Todi-
rhamphus, O-ta-tari by the natives.

Bora Bora, like all of the Society Islands, is volcanic, surrounded
with coral reefs. It is one of the most picturesque islands in Eastern
Polynesia, with its range of craggy peaks, one of which rises to a height
of 2,400 feet, and its encircling barrier reef dotted with cocoanut groves.
It is heavily forested and has a population of about 600.

The Cook or Hervey Isl.\nds. — This group has nine rather
widely separated islands, some of which are volcanic and rise to con-
siderable heights. Others of less height are elevated coralliferous
islands. All are forested and most of them are surrounded by barrier
reefs, often enclosing lagoons.

The group has a native population of over 7,000. Six species of land
birds have long been known, four of which are autochthonous.

Our stop of half a day at Aitutaki, the only one visited, on 21
November, afforded little time for bird collecting. The only bird
found was the Blue lory {Choriphilus peruvianus), of which I secured
seven specimens. These were not difterent from those obtained at
Rangiroa and Bora Bora. This common pet of the natives will
probably be found on several other islands. Aitutaki has a length of
four miles and a height of 450 feet. It is well wooded and watered
and has a population of 1,500.

The Isolated Island of Xiue. Nine or Savage Island, where we
anchored for a few hours on 25 November, is an isolated coralliferous
peak lying 600 miles west of Aitutaki and 250 miles east of the nearest
part of the Tonga Group. Our soundings between Aitutaki and Niue
revealed ocean depths exceeding 2,800 fathoms, and between Niue
and the Tongas, depths exceeding 4,500 fathoms. It has a diameter
of about ten miles, an elevation of 200 feet, and is well wooded. We
saw trees perhaps sixty feet high. The island is fertile and has many
breadfruit and mango trees. It has 4,000 inhabitants. The land
birds secured were fruit pigeon {PtHopus porphyraceus), parrot {Vini
australis), cuckoo shrike (Lalage whitmeei), tree starling {Aplonis
brunnescens), and a White-rumped swift {CoUocalia francica town-
send i), recently described by Oberholser from Albatross collections
as new. This species was also taken in the Tonga Islands. The
genus is celebrated for the nest it builds against clift's, from a secretion
of its salivary glands, and which is marketed as "edible birds' nest."

160 bulletin: museum of comparative zoology.

It was observed at many points during the voyage. The cuckoo
shrike (Lalage) and the tree starling (Aplonis) are both common at
Niue.

This island probably has other land birds, as my brief search for
birds was necessarily confined to the vicinity of Alofi village.

The Tonga or Friendly ^Archipelago consists of about 150 islands
and islets, most of which are volcanic, others being of elevated coral-
liferous limestone. The group is over 400 miles in length. There
are small volcanic islands extending along the western side for about
200 miles, some of which are occasionally active and for that reason
are uninhabited. The three most active volcanoes, Tofua, Kao, and
Late, rise to heights of 1,700 to 3,000 feet.

In the Tongas we found land birds more numerous than in any of
the island groups previously visited.

EuA, (Middleburg), the most southerly of this group, was visited
28 November. During the afternoon in the vicinity of Ohonua village,
I secured six species, fruit pigeon (Ptilopus porphiraccus), cuckoo
shrike {Lalage pacifica), tree starling (Aplonis tahuensis), honey eater
(Meliphaga carunculata), swift (Collocalia francica townsendi), previ-
ously found at Niue, and kingfisher {Sauropatis sacra rabulata), new.
A few other species were observed, but there was no time to search
farther inland. Eua Island is ten miles long, 1,000 feet high, and is
well wooded.

Fruit pigeons (Ptilopus porphyraceus) are abundant in the Tongas.
Numbers could be shot early in the morning, on the wing, as they
approached the high "berry trees" on the fruit of which they feed
habitually. In Niue and the Tongas pigeons in general are known by
the name Kulu-kulu.

ToNGATABU, (New Amsterdam of the older charts), visited 29
November to 1 December, is the largest of the Tongas, with a length
of eighteen miles and an elevation at one part of 200 feet. It is a
rather level island in general and is partly wooded. It lies about ten
miles from Eua. Some of the birds taken here were of the same
species as those from Eua : — Lalage pacifica, Ptilopus porphyraceus,
and Meliphaga carunculata. Another pigeon, Ptilopus perousii, was
added to our list. The kingfisher Sauropatis sacra sacra was different
from that of Eua.

Nomuka, (Annamooka). lying sixty miles north of Tongatabu, is
only two miles in length and has on one side an elevation of 160 feet.
It is wooded and has a small lagoon. Our stop for a short time on
the afternoon of 2 December resulted in the following additions to the

TOWNSEND AND WETMORE: THE BIRDS. 161

bird collection: — Ptilopus porphyraceus, Sauropatis sacra sacra,
Hypurolepis tahitica, Lalage pacifica, Aplonis tahuensis, Meliphaga
carunculata, wood shrike {Pinarolestes heinei), and Polynesian duck
{Anas supercUiosus pelewensis).

Vavau, the last island in the Tongas visited by the Expedition, is
100 miles north of Nomuka. It is nine miles in length and in one part
about 700 feet high. It is cultivated, wooded, and has a population
of over 3,000. The ship lay off Neiafu village 4 and 5 December.
The name Neiafu should not be confused with Niuafou, the latter
being an island about 200 miles north of Vavau. The birds of Niuafou
were made known by Finsch in 1877. It was not visited by the
Albatross. The birds taken at Vavau were Ptilopus porphyraceus,
Sauropatis sacra celada (new), CoUocalia francica townsendi, Lalage
pacifica, a shrike (Pachycephala jacquinoti) , and Aplonis tahuensis.

Wild ducks were seen in the ponds. This island would probably
repay careful ornithological exploration.

A single specimen of Plumalis dominicus fuhus was taken at Vavau.

The Fiji Islands. This great archipelago contains about 150
islands and as many more islets and reefs. It extends through seven
degrees of latitude and of longitude. The islands are of both volcanic
and of coral formation, are densely forested, and have a native popula-
tion of more than 100,000.

Kambara, where the Albatross lay from 7 to 9 December, is a
wooded, volcanic island, four miles long, with a height of 470 feet.

Two of the birds found in the Tongas, Ptilopus porphyraceus and
Lalage pacifica, were also found here. The other land birds were
Aplonis vitiensis, honey eater {Myzomcla jugularis), Halcyon sacra
vitiensis, flycatcher {Myiagra townsendi), and wood shrike {Pinaro-
lestes nesiotes), the last two being new species.

The honey eater, Myzomela, was the only species that appeared to
be common near the village of Tokalau, where all the birds were
obtained. Its native name is Bithi-bithi Kula. The new flycatcher,
Myiagra, is called Sia-sia. Other birds might have been found in the
higher woods, had there been time to search for them. There are
apparently no records to show that Kambara had ever been visited
by naturalists. There are perhaps a dozen islands in the Fiji Group,
of the size of Kambara, about which naturalists have no information.

The Albatross arrived at Suva, Viti Levu Island, 10 December,
and remained until the 19th. Viti Levu, largest of the Fijis, is also
the largest island of Eastern or Central Polynesia, having an area of
over 4,000 square miles. Its highest peak has an elevation of 5,000

162 bulletin: museum of comparative zoology.

feet. The rainfall is heavy and the largest river is navigable for small
vessels for forty miles. It is heavily forested and the climate is that
of the moist tropics. Parts of Viti Levu are well cultivated and the
population is large.

Work pertaining to the fisheries and to ethnological collecting pre-
vented systematic bird collecting in Viti Levu, but the following were
secured: — Lalage pacifica, Myzomela jugularis, Meliphaga procerior,
Zosterops fiaviceps, Acridotheres tristis, and flycatcher {Haplornis
lessoni) .

The Mynah {Acridotheres tristis) is common. I did not ascertain
when it was introduced. I found it abundant in the Hawaiian Islands
twenty-five years ago. It is also common in Tahiti. Wherever
introduced it becomes a menace to the native island species. The
same may be said of the Mongoose now common on Viti Levu.

I was detached from the expedition at Suva and returned home by
way of Samoa, the Albatross proceeding northwestward through the
Ellice, Gilbert, Marshall, Caroline, and Ladrone Islands to Japan.

The Samoan Islands. The birds picked up at Apia, LTpolu
Island, 26 December, were a parrot {Vini auMralis), kingfisher
{Todirhamphiis recurrirostris), honey eater (Myzomela nigripentris)
and Meliphaga carunculata. The last is rather widely distributed,
having been previously taken at several points in the Tongas.

Dr. H. F. Moore took up the work of bird collecting after I left the
Albatross in the Fijis. He found the bird life of the Caroline Islands
richer and more varied than in an;^' of the groups visited during the
voyage : —

"In the Ellice, Gilbert, and Marshall islands land birds are extremely
uncommon and of but few species, the avifauna being poorer than in the
Paumotus. The Society and Fiji Islands are progressively richer, but it was
not until the Carolines were reached that the woods and thickets seemed full
of birds and resounded with their songs and cries. Parrots and pigeons of
several species, white-eyes, flycatchers, kingfishers, and many other species
were observed at Kusaie, Ponape, and Truk, and the collections, which, in
spite of effort, had languished for lack of material after leaving Suva, began to
offer some returns to the shooters notwithstanding the brevity of the oppor-
tunities, which made it impossible to secure a really representative collection."

The Ellice Islands, extending in a northwesterly direction for 360
miles, are low atolls, most of them with central lagoons. Funafuti,
the onlv one from which birds were taken, is an atoll thirteen miles
long. It was visited 23 December. The land birds were the large

TOWNSEND AND WETMORE : THE BIRDS. 163

pigeon {GJobicera pacifica), and a cuckoo {Urodynamis taitensis taiten-
sis), said to lay its eggs in the nests of the noddy terns. These are
the only land birds known to this group of islands. The water birds
were PlimaUs dominicus fiihus, Hctcraditis incanus, Limosa lapponica
baueri, and the noddy {Anoiis sfolidus pileatus).

The Gilbert Islands, having about the same extent as the EUice
Islands, are also atolls. Land birds were not obtained, but the follow-
ing water birds were taken at Taritari (Butaritari) 6 January: Are-
naria interprcs oahucnsis, Phacopus tahiticnsis, Hcteractitis incanus,
Plnvialis dominicus fidvus, and Pisobia acuminata.

The Marshall Islands. The Albatross cruised among the low
atolls of the Marshall Islands from 9 January to 5 February. Land
birds were not obtained. The water birds were the same species as
those taken in the Gilberts with the exception of Sterna sumatrana
from Arnho Atoll, 24 January. Two land birds are known to the
Marshall Islands, Urodynamis taitensis GJobicera and oceanica.

The Caroline Islands. The high volcanic islands of the Caroline
Archipelago proved to be rich ground for bird collecting after a long
cruise among the ornithologically barren atolls.

The Albatross was at Kusaie (Ualan, Strong) from 7 to 9 Febru-
ary. This is a volcanic island twenty -four miles in circumference and
over 2,000 feet high. It is heavily forested and well watered. The
land birds taken were Ptilopus hernsheimi, Aplonis opaca, Myzomela
rubratra nibratra, Zosterops cincrea, and GJobicera oceanica oceanica.
The water birds were Demigrctfa sacra, Hcteractitis incanvs and Anous
stoJidus piJeatus. About nine species of land birds were previously
known to inhabit Kusaie.

PoNAPE, or Ascension Island, was visited 11 and 12 February. It
is a volcanic island which, with its surrounding coral reef, has a diame-
ter of about seventeen miles. It has a height of nearly 3,000 feet, is
heavily forested and well watered. The land birds taken were
Zosterops ponapencnsis, ApJonis opaca, MyzomeJa rubratra dichromata
(new), Conopoderas syrinx, Myiagra pJuto, Rhipidura Jcubaryi, Sauro-
patis mcdiocris, Eos nihiginosa, and GJobicera oceanica townsendi (new).
About eighteen species of land birds were known to Ponape when
reported upon by Finsch in ISSO.

Uala, or Moen, is one of the small but lofty volcanic islands known
as the Truk, Ruk or Hogelu Group all lying within a great lagoon.
Truk is the largest atoll of the Carolines, the circumference of the
lagoon enclosed by the outer barrier reef being 125 miles. L^ala is
1,300 feet high and Ruk 1,000, while several of the others are nearly

164 bulletin: museum of compara.tive zoology.

as high. All of the islands are wooded and have a total population of
several thousands. The Albatross anchored at Uala 14-17 February,
where the following birds were collected : — Ptilopus ponaj^ensis,
Metaholus rugensis, Myiagra oceanica, Conopoderas syrinx, Zosterops
semperi owstoni, Aplonis opaca, and Myzomela rubrata riibrata.
Water birds secured were Nycticorax cahdonicvs, Ixobrychus situmsis
moorei (new), Pluvialis dominicus fulvus, Arenaria inter pres oahuensis,
and Hderactitis brevipe^.

The adjacent islands of the lagoon were not visited. One of them.
Ruk, had eleven species of land birds when reported upon in 1900.

The Ladrone Islands. Guam, visited 21 to 25 February, was
^ the last island at which birds were collected during the cruise. This

island is partly volcanic and partly elevated coralliferous limestone.
It is twenty-two miles in length, forested, and has a height of 1,000
feet. The only land bird collected was the quail (Excalfadoria chinen-
sis lineata) introduced from the Philippines. The water birds ob-
tained were Phaeopus phaeopus variegalus, Gallinula chloropns, and
Ixobrychus sinensis bryani. Twenty-seven species of land birds are
known to the island of Guam. The expedition proceeded northward
through the Ladrone or Mariana Islands, arriving at Yokohoraa 4
March, 1900.

Part 2. — Annotated List of the Species. By Alexander Wetmore.

The collection of birds made during the cruise of the Albatross
in Polynesia during the winter of 1899-1900 numbers 391 skins,
fourteen alcoholic specimens, and one skeleton. These specimens
represent ninety-three distinct forms, of which fourteen are here
described for the first time.^ Collections of birds were made on
thirty-three islands some of which were little known and were visited
by an ornithologist for the first time. The entire collection has
considerable general interest, as a number of species were collected
at their type-localities, and in several cases important series of such
birds were secured. In working out this rich material I have been
under deep obligation to Dr. C. \Y. Richmond, Associate Curator of
Birds in the U. S. N. M., for advice and assistance in matters pertain-
ing to bibliography and nomenclature. At one time Dr. Richmond
had planned to publish on this collection personally but was prevented

1 Three new species of CoUocalia collected during this cruise of the Albatross were described
by Oberholser in 1906, see p. 200-201.

TOWNSEND AND WETMORE : THE BIRDS. 165

from carrying out his intention by press of other work. In completing
the identification of these specimens he permitted the use of manu-
script notes, made during his preUminary examination, tliat have
proved of much value.

Much difficulty has been encountered in identifying the subspecific
forms in many species discussed in the.following pages through lack of
sufficient material for comparison. These cases have been treated
with as much care and attendon as practicable, but in some instances
it has been impossible to assign anything more than a specific name.
In such comparisons the early collections of Titian R. Peale, made
during the U. S. Exploring Expedition of 1838-1842, have been of
great value. The importance of the birds in Peale's collections from
an historical standpoint is not to be over-estimated, especially as a
large part of the original type-specimens upon which Peale based his
names of new species have been available for study. This collection
was formerly mounted, but now all of the birds have been taken down
and remade and are kept as study skins.

The treatment of the fruit pigeons of the genus Ptilopus in the
present paper is highly unsatisfactory. The entire group of these
pigeons is badly in need of revision, but in the present case this was
impracticable because of a lack of sufficient material from many
localities. Little attempt has been made here to accord the species
any other treatment than that given them by Count Salvadori in the
twenty first volume of the British Museum Catalogue of Birds, but
it is believed that a number of new forms may be described eventually
from the specimens listed herein. Additional collections from other
islands are needed, however, before these may be diagnosed properly.

A series of 105 skins, forty-seven species or subspecies, from the
specimens collected during the Albatross expedition, has been placed
in the M. C. Z. The remaining specimens, including the types of
forms described here as new, are in the collections of the U. S. N. M.

In the following pages is given an annotated list of the species and
subspecies identified, with descriptions of such forms as appear to be
new. All measurements are given in millimeters.

166 bulletin: museum of comparative zoology.

PHAETHONTIDAE.

1. Phaethon rubricaudus Boddaert.

Phaeton rubricauda Boddaert, Tabl. plan, enl., 1783, p. 57. (Mauritius).

Three specimens were collected at sea about 600 miles east of the
Marquesas Islands in Latitude 10° N., Longitude, 130° W., 2 Septem-
ber, 1899. All three are immature. One has the upper parts heavily
barred with black, but in the other two these markings are less evident.
The red-tailed tropic-birds without doubt are divisible into two or
more subspecies, but in the absence of material from many localities
needed for a competent review of the forms no attempt is made to
allocate the present specimens subspecifically. The fact that all are
immature would add to the difficulty of such an identification. The
measurements of these birds are as follows : —

No.

Sex

Wing

Tail

Culmen

Tarsus

U. S.N. M. 212,164,

.31.3

232

212,16.5,

314

201

28.5

M. C. Z. 81,927 ( " 212,166)

301

190

29.5

These skins are similar in size to specimens from Laysan Island, in
the Hawaiian Group.

It has been proposed by Mathews (Austr. avian record, 1913, 2,
p. 56) to separate the Red-tailed tropic-bird from Phaethon under the
generic name Scaeophaethon, on the grounds that it has a longer wing,
stronger legs and feet, and shorter tail. Upon careful comparison
it is found that all of these characters do not hold. Thus when
compared with Phaethon aetherevs, the type-species of the Linnaean
genus Phaethon, specimens of P. rubricaudus from Assumption and
Gloriosa Islands (north of Madagascar) do have the wings longer.
On the other hand red-tailed birds from Laysan Island have the wing
equal to or shorter than that of the red-billed species. In other words
birds from these two localities, representing only well-marked forms
of one species, would be placed in different genera on this character.
The feet and tarsi are slightly stronger and the tail is shorter in
riihricandtis when series of the two species are compared. There is
so much individual variation in respect to length of tail that it is of
value only as an average character. In addition there are other

TOWNSEND AND WETMORE : THE BIRDS. 167

structural characters separating the two that Mr. Mathews over-
looked. From the series available it seems that Scaeophaethon has
the operculum over the nostril broader and heavier, and barbs on the
shafts of the two elongate rectrices on either side greatly reduced in
length. This latter character is one by which Scaeophaethon may be
recognized at a glance. The black line at the side of the black shaft in
P. rubricmidus is misleading as it makes the shaft appear broad and
strong while in reality it is the same size as the shaft in P. aether eus.
When old and much worn the central rectrices of P. aethcreus become
narrowed and resemble those of Scaeophaethon but may be distin-
guished by their ragged appearance. As the small-billed P. americanus
also has a strong broad operculum the basis of differentiation falls
upon differences in the tail alone. These are assumed here to be only
subgeneric in value and the Red-tailed tropic-bird is kept in Phaethon.

SULIDAE.
2. SuLA PiscATOR (Linne).
Pelecanus piscator Linne, Syst. nat., ed. 10, 1758, 1, p. 13-4. (Java Seas).

Three specimens of this booby were collected at Tekokoto in the
Paumotu Islands; 26 October, 1899. None of these is in adult plumage
though all are one year old or more. One specimen, a male, has the
tail and head white, while the back and lesser wing coverts are hair-
brown. In the two remaining specimens the tips of the rectrices are
white and the rest of the plumage is dull.

Mathews (Birds of Australia, 1915, 4, pt. 3, p. 216) states that
specimens of this booby from the Pacific Ocean are larger in every
dimension than those from the Atlantic region and that " the soft
parts seem to differ." For this reason he separates the Australian
bird under the subspecific name ruhripcs Gould. Concerning the
color of the soft parts information of value is not available, but com-
parison of a series of skins from Pacific and Atlantic Ocean localities
fails to substantiate the claim made as to dift'erence in size. So far
as measurements of wing, culmen, and tarsus go, specimens from the
Paumotu Islands are almost identical with birds collected by the
author on Desecheo Island, a small island lying between Porto Rico
and Santo Domingo in the West Indies. Careful study of a larger

168 bulletin: museum of comparative zoology.

series than that at hand will probably show that birds from the
different ocean areas may be separated as subspecies, but for the
present it is thought best to use the specific name for the birds in hand
without attempt at subdivision. Especially this is the case since no
Australian specimens are available for comparison.

The question of the separation of the Sulidae into genera is one that
is subject to individual opinion more or less. There is no question
that the three species of gannets form a well-characterized genus;
but that there are trenchant lines separating the smaller species known
as the boobies into groups that may be considered of generic rank,
seems at present uncertain. The differences indicated rather signify
only subgeneric differences. For the present it is proposed to ignore
them and to include all of the smaller Sulidae in Sula, pending further
study of available material that may throw light on the subject from
another angle.

Recently Mathews (Birds of Australia, 1915, 4, pt. 3, p. 212) has
replaced Pelecanus piscator Linne, the name in common use for the
Red-footed booby, by Pelecanus sula Linne, on the grounds that
Pelecanus piscator, as used by Linne, was a name based upon a com-
posite species, and that the form to which it properly belonged was
indeterminate. The original name, however, must stand, as the fol-
lowing will show.

The name Pelecanus piscator appears in the tenth edition (1758) of
Linne's Systema naturae on page 134. The description there is
meager, and, as Mathews has shown part of the references there given
are indeterminate, while a part belong properly to the bird known at
the present time as Sula leucogastra (Boddaert). This, however,
does not hold for all the citations noted. The first reference is,
literally transcribed, "Chin. Lagerstr. 8." Mr. Mathews cites this
but evidently did not have the work available, and so was forced to
base his argument upon the second reference to " Osbeck iter, 85."
The paper cited as " Chin. Lagerstr." is an inaugural dissertation
entitled Chinensia Lagers tromiana by John L. Odhelius. From Dr.
C. W. Richmond it is learned that this was printed first, as a separate
publication of 36 pages, in 1754. Later in 1759 it was reprinted as
number 4 in a collection of inaugural dissertations known as the
Amoenitates academicae. A. copy of this reprint is available and on
reference it is found that species number 8 is given as "Pelecanus
{piscator)." The pertinent portion of this reference is quoted here in
full, as the tract in question is rare and not to be found save in large
libraries : —

TOWNSEND AND WETMORE : THE BIRDS. 169

"8. Pelecanus (piscator) rostro serrato, cauda cuneiformi.

A. Anseri bassano affinis fusca avio. Sloane jam.

B. Anseri bassano congener cinereo-albus. Sloane jam.
prsef. 31. t. 6. f. 1. Raj. aves 191.

Bubbi chinensibus.

Hujus duo adsunt sexus.

MAS (a) totus niger, abdomine canescente.

FEMINA (/?) tot a albida, remigibiis nigris.

Rostrum utrisque gibbum, in foemina pra>cipue sanguineum,

margine tenuissime retrorsum

serrato.
Gula nigra.

Corpus magnitudine anatis ma j oris.
Pedes sanguinei, magni, tetradactyli, digitis omnibus com-

muni membrana junctis; unguis intermedii margo interior

dilatatus & fere pectinatus.
Aloe utrius que sexus subtus albicant.
Redrices caudae XIV, interioribus sensim longioribus, in

foemina etjam parum fuscescentibus."

There is little question that the male and female described above
belong to separate species of which the female is the bird now known
as Sula piscator.

Linne himself recognized that this name covered a mixture of two
species, and in his twelfth edition of the Systema naturae (1766, p.
217) he again gives Pelecmms piscator with a slightly different diagno-
sis, and the reference "Amoen. acad. 4, p. 239. femina." This cita-
tion refers to the reprint published in 1759, on page 239 of which is
found the description as quoted above. Linne as first reviser of the
species has here restricted the name Pelecanus piscator to the female
of the bird described by Odhelius, and there can be no doubt but that
the Red-footed booby is intended. Mathews objects to Linne's
statement that the flight feathers are black on the grounds that in
the Red-footed booby the outer webs of the quills have a hoary gray
appearance. This is true, but at the same time the body color of the
feather is black, and to a casual inspection the entire feather appears
blackish. The older naturalists were not so critical of color differences
as are ornithologists today, so that we may overlook this slight error
as the rest of the description tallies closely. Because of this state-
ment that the wing feathers are black Mr. Mathews suggests that the
bird described was Sula abbotti Ridgway, a species with intensely
black flight feathers. This cannot be true, however, as Odhelius

170 bulletin: museum of comparative zoology.

states that in his Pelccanus jnscator there are fourteen rectrices while
the type of Sula ahhotti (the only specimen available) possesses sixteen.
The name Pelecanvs ■piscator Linne, therefore, is still available for the
Red-footed booby.

3. Sula leucogastra plotus (Forster).

Pelecamis plotus Forster, Descrip. anim., 1844, p. 278. (Near New Caledonia).

An adult bird was taken on Tekokoto, Paumotu Islands, 26
October, and another was preserved as a skeleton from Mehetia, in
the Society Islands during November, 1899. The skin from Tekokoto,
with other birds examined from the Hawaiian Group and elsewhere
in the Pacific, agrees with Mathews's description of the Australian
form (Birds of Australia, 1915, 4, pt. 3, p. 234) and differs from birds
from the Atlantic region, in darker coloration above and in being
slightly larger in size. No specimens have been examined from
Australia in the present connection, but it is assumed that they are
the same as the bird from the Paumotu Group. The difference in
color between these birds, and those from localities in the Atlantic
Ocean is well marked, and the races thus indicated seem to be well
defined.

FREGATIDAE.

4. Fregata minor palmerstoni (Gmelin).

Pelecamis palmerstoni Gmelin, Syst. nat., 1789, 1, pt. 2, p. 573. (Palmerston
Island).

One specimen was taken, a female labeled "Polynesia." This bird
has the throat and breast white, and the abdomen black. The culmen
measures 117 mm., the wing 595 mm. A specimen in the U. S. N. M.
from Kaui and two others from Laysan Island, in the Hawaiian
Group, have the feathers of the "wing bar" with paler edgings than
in this bird, and with a larger series it may be possible to recognize
the form named strumosa by Hartert, as Mathews has done. In
addition these three northern birds have a metallic sheen on the
feathers of the back which is lacldng in the specimen from Polynesia.

TOWNSEND AND WETMORE: THE BIRDS. 171

5. Fregata ARIEL ARIEL (G. R. Gray).

Atagen ariel Gray, Gen. birds, 1845, 3, plate 183. (Raine Island, North
Australia).

One specimen, a male, was taken at Makemo, Paumotu Islands,
25 October, 1899. The culmen measures 86.5 mm. and the wing
495 mm. The gular pouch is not at all developed. There are so few
specimens of this frigate-bird available that it is difficult to make out
the forms into which it may properly be di^dded. The bird in hand
is referred to the typical form.

ARDEIDAE.
6. Demigretta sacra sacra (Gmelin).

Ardea sacra Gmelin, Syst. nat., 1789, 1, pt. 2, p. 640. (Tahiti).

Seven specimens were collected in the Paumotu Islands, the Gilbert
Islands, and Kusaie in the Eastern Carolines. Two males from
Makemo in the Paumotu Islands, were collected, 20 and 23 October,
1899, respectively. One is in fine dark plumage, while the other is
white save for the elongate dorsal plumes. Of two birds collected at
Rangiroa on 21 September, one sexed questionably as a female is also
white with dark markings on the longer feathers of the back, while the
other (a female) has the crown, sides of neck, breast, wings, and tail
pied with dark markings, with white as the predominant color. Two
males were taken at Tarawa, Gilbert Islands, 3 January, 1900; one
is entirely white, while the other is pied as described in one bird from
Rangiroa. A male from Kusaie," collected 8 February, 1900, has one
dark feather among the right scapulars, but is white elsewhere. This
specimen and the white bird from Tarawa are remarkable in having
the elongate dorsal plumes white.

The status of the white and dark birds is somewhat uncertain, some
ornithologists considering them to be distinct species. As dichro-
matism is so well known among other herons it is only reasonable to
suppose, however, that these birds represent a similar case in the reef-
herons. There are evident no structural characters by which light
and dark birds may be separated, and color in this instance must be

172 bulletin: museum of comparative zoology.

considered wholly unreliable, as is shown by the description given
above of the specimens in hand. For the present these are referred
to the typical subspecies.

7. Nycticorax caledonicus (Gmelin).
Ardea caledonica Gmelin, Syst. nat., 1789, 1, pt. 2, p. 626. (New Caledonia).

An adult female was taken at Uala in the Middle Carolines, 16
February, 1900. This bird is seemingly in fully adult plumage but
lacks the long white nuchal plumes found in this species when in full
nuptial dress.

It is darker above than birds from New South Wales (A^. c. hilli
Mathews), Waigou, and New Guinea, and has the axillars and under
wing coverts heavily washed with buff-pink, a character lacking in the
few other specimens examined, though said to be found in some
Australian birds. A dark purplish wash on the back is especially
noticeable in this specimen when compared with others and the bill
seems thick and heavy. The measurements of this bird are as follows :
— wing 280 mm., tail 97 mm., tarsus 80 mm., exposed culmen 62 mm.
The length of culmen is uncertain as the bill seems to have sustained
some injury near the base of the culmen that has caused distorted
feathers to come farther forward on the forehead than usual. On the
right foot this specimen had lost all of the phalanges of the middle toe
save the basal one and the nail from the second toe, leaving only well-
healed stumps at the tips of these digits.

Apparently the Caroline Island bird represents a form characterized
by dark coloration above, a pinkish wash on the under wing coverts,
and a thick rather short bill. The short bill serves to separate it from
N. c. crassirostris from the Bonin Islands, and the dark coloration
from A^. c. hilli Mathews from Australia. No material from New
Caledonia, the type-locality of caledonicus, is available, so that I find
myself unable to definitely name or differentiate the Caroline form.

The bill in the present specimen, as in all others that have been
examined, has the basal portion of the mandible yellowish and the tip
of the mandible and the maxilla black. It is said that A^. caledonicus
at times has the entire bill black as in A'^. manillensis Vigors, but I have
seen none that exhibit this character. Mathews (Birds of Australia.
1914, 3, pt. 6, p. 460) says, in -his description of A^. c. hilli, that the
bill is black, and it is so figured in the plate that he gives of this
night heron.

TOWNSEND AND WETMORE: THE BIRDS. 173

8. IXOBRYCHUS SINENSIS BRYANI (Seale).

Ardeita bryani Seale, Occas. papers Bernice Pauahi Bishop mus., 1900, 1,
no. 3, p. 27. (Guam).

An immature male of the Little yellow bittern was taken on Guam,
24 February, 1900. Comparison of a large series of these small
bitterns in the U. S. N. M. from various localities (including adult
birds taken on Guam) shows that the bird described by Seale as
Ardctta hryani may be recognized as a valid form of the widespread
Ixohryckus sinensis.

9. IXOBRYCHUS SINENSIS MOOREI, subsp. nOV.

Characters. — Similar to Ixohryckus sinensis hryani (Seale) from
Guam but back darker and duller in color, more grayish; distal portion
of scapulars and tertials duller, and grayer; sides of head and neck
much more pinkish, this color sharply defined from buff of throat, and
not merging gradually into it; sides of neck much brighter, more
pinkish brown.

Type.— V. S. N. M. 212,171. Adult male. Polynesia: Middle
Caroline Islands; Truk Group, Uala, 16 February, 1900. H. F. Moore.

Description. — Feathers of crown dusky neutral gray, changing to
deep neutral gray at sides and on elongate crest, a slight brownish
wash evident anteriorly; feathers of hind neck cameo-brown, this
color extending to shoulders; back bister, with a slight intermixture
of snuff-brown; rump deep mouse-gTay, shading into blackish mouse-
gray at tips of upper tail coverts; elongate scapulars between snuff-
brown and bister; tertials mouse-gray, the longer ones washed at the
tips with snuff -brown; primaries black, the outer one margined
indistinctly with paler, especially near tips, the others washed some-
what with neutral gray; outer secondaries black, washed lightly with
neutral gray, inner ones snuff -brown; greater, middle, and lower
lesser coverts between honey-yellow and isabella-color, with a slight
wash of warm buff"; shoulder and upper lesser coverts snuff-brown;
large feathers of alula dusky neutral gray, the outer one margined
broadly with light buff; anterior margin of wing white with a faint
buffy tinge; sides of head and neck between mikado-brown and
verona-brown, this color sharply differentiated from lighter color of
throat and neck, somewhat less sharply demarked posteriorly ; throat
white with a very faint tinge of buff; foreneck pinkish buff, becoming

174 bulletin: museum of comparative zoology.

nearly white toward upper breast; a dark patch on either side of
upper breast, that is almost concealed by elongate feathers of neck,
fuscous black above changing to deep mouse-gray below, the feathers
all broadly margined with pinkish buff; rest of breast and sides dull
cream-buff; abdomen and under tail coverts white, very faintly
washed with buff; flanks honey -yellow; axillars and under wing
coverts white, faintly tinged with buff; a small area of honey-yellow
at bend of wing on under side. Distal half of culmen brownish black;
rest of maxilla, except basal part of tomia, dull dark brown; base of
mandibles, gonys, and maxillar tomia at base pale dull brownish buff;
tip of mandible and sides dark dull brown; loral space dull brown;
eye ring somewhat paler; tarsus and toes dark, dull brown, claws
darker (from dried skin).

Measurements. — Male adult (Type) wing 128.5; tail 44; culmen from
base 53.5; tarsus 45.2.

Range. — Island of Uala, Truk Group, Middle Caroline Islands,
Polynesia.

Remarks. — This subspecies is based on a single specimen taken on
the island of Uala. The differences noted in this one bird are not
approached by individual variation in the large series of little yellow
bitterns examined, so that there is no question but that the divergence
shown is of subspecific value. In size the type of the new form char-
acterized here is slightly larger than Lvobryckus sinensis astrologus
Wetmore (Proc. Biol. soc. Washington, 191S, 31, p. 83) recently
described from the Philippine Islands, and, like /. s. hryani from
Guam, it resembles the Philippine Island bird in color. It may be
distinguished from /. s. astrologus by being darker, grayer, less rufes-
cent on the back, by having the tertials and scapulars duller and less
brownish, the under tail coverts whiter, and the sides of head and neck
brighter brown, more sharply defined from the color of the median
line.

From the material at hand it is possible now to recognize the follow-
ing races of the little yellow bittern : —

IxoBRYCHUS SINENSIS SINENSIS (Gmelin) (Ardea sinensis Gmelin,
Syst. nat., 1789, 1, pt. 2, p. 642. China). China (Tientsin, Hankow,
Tung Chow, near Hongkong).

IxoBRYCHUS SINENSIS LUTEOLUS (Stejuegcr) {Ardetla lideola Stej-
neger, Proc. U. S. N. M., 1888, 10, p. 289. Wakayama, Kii, Hondo)
Japan (Islands of Yezo, Nippon, and Hondo).

IxoBRYCHUS SINENSIS LEPIDUS (Horsfield) {Avdea Jepida Horsfield,
Trans. Linn. soc. London, 1821, 13, p. 190. Java). Java (Sumatra?).

TOWNSEND AND WETMORE : THE BIRDS. 175

IxoBRYCHUS SINENSIS ASTRO LOGUS Wetmore (Proc. Biol. soc.
Washington, 1918, 31, p. 83. Luzon). Philippine Islands.

IxoBRYCHUS SINENSIS BRYANi (Seale) {Ardctta hryani Scale, Oceas.
papers Bernice Pauahi Bishop mus., 1901, 1, no. 3, p. 27 Guam).
Guam, Marianne Islands.

IxoBRYCHUS SINENSIS MOOREi Wetmore, Supra, p. 173, Uala,
Middle Caroline Islands.

In the series examined other localities than those listed above are
represented only by birds in immature plumage whose subspecific
identification is somewhat uncertain until additional material is
available. There are without ciuestion several other races present,
so that the forms listed are gi^'en merely to render the treatment
accorded the Caroline Island bird intelligible. It may be noted that
another name is available for a form of this bittern as the bird from
the Andaman Islands has been described by Hume (Stray feathers,
1873, 1, p. 309) as (Ardctta) pulchra, but no specimens from that
locality have been available in the present study.

ANATIDAE.

10. Anas superciliosa pelewensis Hartlaub and Finsch.

Anas superciliosa Gmel. var. pelewensis Hartlaub and Finsch, Proc. Zool.
soc. London, 1872, p. 108. (Pelew Islands).

Two specimens were taken, in the Tonga Islands, a male at Nomuka,
2 December, 1899, and an adult bird with sex not marked, at Vavau,
5 December, 1899. The type-locality of Anas superciliosa Gmelin
is New Zealand and the bird from Australia which is distinctly larger
than the typical form has been separated by Mathews (Austr. avian
record, 1912, 1, p. 33) as A. s. rogcrsi. Polynesian birds (from Upolu,
Samoan Islands, Nomuka, and Vavau, Tonga Islands, and Tahiti)-
are similar in size to the New Zealand form, but differ from both .1. s.
superciliosa and A. s. rogcrsi in being distinctly darker below and in
having the throat, band on side of head, and superciliary stripe
pinkish buff. In superciliosa jDroper and in rogcrsi the buff' on these
areas is decidedly less rufescent. INIeasurements of the two specimens
in the present collection are as follows: —

No. Sex Locality Wing Tail Culmen Tarsus
U. S.N. M. 212,167 o" Nomuka, Tonga Islands 217.5 75.2 48.0 43.0
212,168 ? Vavau, " " 240.0 42.6 44.5

176 bulletin: museum of comparative zoology.

These are placed with some reserve under pelewensis, (no specimens
of which are available) as in general they agree with the original
description of that form.

PHASIANIDAE.

11. Gallus gallus (Linne).
Phasianus gallus Linne, Syst. nat., ed. 10, 1758, 1, p. 158. (India orientali).

An adult male in full plumage was collected on Nukuhiva in the
Marquesas Islands, 16 September. This bird is somewhat larger than
birds from Siam and elsewhere in the proper range of G. gallus, and
has larger feet and heavier tarsi than in the average specimen from
other localities. Peale (U. S. explor. exped., 1848, 8, p. 179) found
the Jungle Fowl wild on the island of Tahiti, and collected several
specimens. He called attention to the fact that his birds were "a
shade lighter in colour, somewhat larger, with broader pendant
feathers in the tail, and larger comb which is entire on the hind part,"
and gave figures illustrating the heads of birds from Tahiti and Ma-
lacca. Dr. Richmond has called my attention to the fact that Hartlaub
(Journ. fiir ornith., 1854, p. 169) named this bird of Peale's Gallus
takitiensis, basing the name on Peale's figure and description. This
name, changed to Gallus tahiticus, was used by Cassin (U. S. explor.
exped., 1858, 8, p. 290) who remarks that "In the collection of the
expedition we find a well-characterized specimen from Tahiti."
From Dr. Richmond it is learned that Baird was unable to find this
bird when making a manuscript list of the birds in the Smithsonian
Museum, between 1860 and 1865, as he wrote "not found" at the top
of the sheet devoted to this specimen.

The specimen examined from Nukuhiva has the back of the comb
entire, but this seems to be a character of little weight as wild birds
. from elsewhere agree with it in this point; it has the following measure-
ments:^— wing 236; tail 206; culmen (from comb) 18.2;' tarsus 89.3;
middle toe with claw 68.5.

PERDICIDAE.

12. EXCALFACTORIA CHINENSIS LINEATA (Scopoli).

Oriolus lineatus Scopoli, Deliciae florae et faunae Insubricae, 1786, pt. 2,
p. 87. (Luzon).

A male was collected at Guam, 24 February, 1900. Scale (Occas.

TOWNSEND AND WETMORE : THE BIRDS. 177

papers Bernice Pauahi Bishop mus., 1901, 1, no. 3, p. 37) states that
this quail was introduced into Guam from Manila in 1894 by Captain
Pedro Duarty of the Spanish Army.

RALLIDAE.

13. Gallinula chloropus (Linne).

Fulica chloropus Linne, Syst. nat., ed. 10, 1758, 1, p. 152. (England). ^

An adult female was collected at Guam, 24 February, 1900. This
bird has the following measurements : —

No. Wing Tail Tarsus

U. S. N. M. 212,180 158.0 62.5 48.0

The material at hand representing this species from localities
outside the United States is too scanty to permit satisfactory study
of subspecies at present, so that this specimen is simply catalogued
as above without attempt to show its subspecific relationships.

CHARADRIIDAE.

14. Pluvialis dominicus fulvus (Gmehn).

Charadrius fulvus Gmelin, Syst. nat., 1789, 1. pt. 2, p. 687. (Tahiti).

Eighteen specimens of the Eastern or Pacific golden plover were
secured from the following localities: — Makatea, 6 October; Tikei,
9 October; Makemo, 20 October, (Paumotu Islands); Vavau, (Tonga
Islands) 4 December; and Arhno Atoll, 25 January (Marshall Islands);
Tarawa, 3 January; Funafuti (EUice Islands) 24 December; Rongelab,
18 January, and Taritari, 6 January (Gilbert Islands); and Uala
(Middle Carolines) 16 February. A female from Tikei has many
black feathers on the under surface. A. male from Makemo is less
strongly marked with black. Others are all in full winter plumage.

ARENARIIDAE.

15. Arenaria interpres oahuensis (Bloxam).

Tringa oahuensis Bloxam, Byron's Voy. Blonde to the Sandwich Islands, 1826,
p. 251. (Sandwich Islands) .

Eleven specimens obtained were collected as follows : — One female

1 Hartert, Nov. zool., 1902, 9, p. 424.

178 bulletin: museum of comparative zoology.

and one specimen without sex, Funafuti, EUice Islands, 24 December,
1899; two males and one female, Tarawa, Gilbert Islands, 3 January,
1900; a male and a female, Taritari, Gilbert Islands, 6 January, 1900;
a male, Jaluit, Marshall Islands, 10 January, a male and an unsexed
specimen, Rongelab, Marshall Islands, 18 January, and a female at
Uala in the Middle Carolines on 16 February. All of these birds are
in full winter plumage. Comparison of a small series of European
birds shows that the Pacific turnstone when in breeding plumage
differs in having the chestnut areas of the back slightly darker.

SCOLOPACIDAE.

16. PiL\E0Pus phaeopus variegatus (Scopoh).

Tantalus variegatus Scopoli, Deliciae florae et faunae Insubricae, 1786, pt. 2,
p. 92. (Luzon).

A male was collected at Guam in the Ladrone Islands, 24 February,
1900.

Mathews (Birds of Austraha, 1913, 3, pt. 2, p. 168-169, 175) states
that the species included in Phaeopus diflfer from the three species
allotted to true Numenius {N . arquata, cyanopns, and americamis)
in having the bill shorter than one half the wing, shorter than the tail,
shorter than the tarsus and middle toe together, the tail longer than
the tarsus and middle toe, and the middle toe more than half the tar-
sus. Careful comparison of all of the species involved shows that
Numenius differs structurally from Phaeopus only in ha\ing the bill
longer than the tarsus with middle toe, and longer than the tail. In
Phaeopus the bill is shorter than the tarsus with the middle toe, and
equal to or shorter than the tail. Though in most Numenius the bill
is longer than half the wing, in some of the adult specimens of all
three species included here the bill is less than one half the wing, as it
is in all the forms belonging under Phaeopus. Though the tail is
shorter than the tarsus with middle toe in all three species referred to
Numenius, it is also shorter in Phaeopus takitiensis, though longer
in all the other species of Phaeopus. The length of the middle toe
compared with the length of tarsus is found to be more in Phaeopus
but variable in Numenius, so that it has no value as a generic char-
acter. The valid structural differences between the two genera may
be summed up as follows: —
a. Bill longer than tarsus with middle toe, longer than tail.

Numenius.

TOWNSEND AND WETMOKE : THE BIRDS. 179

aa. Bill shorter than tarsus with middle toe, equal to or shorter than

tail Phaeopus.

The species of Numenius are larger than those of Phaeopus, but
there is no pronounced gap between the two groups, as P. tahiticnsis
and P. hudsonicus form intermediate steps between the larger and
smaller curlews.

17. Phaeopus tahitiensis (Gmelin).

Scolopax tahitiensis Gmelin, Syst. nat., 1789, 1, pt. 2, p. 6.56. (Tahiti).

Ten specimens of this fine curlew were collected as follows : — Maka-
tea, 6 October, and IVIakemo, 22 October, Paumotu Islands; Taritari,
Gilbert Islands, 6 -lanuary; and Rongelab, Marshall Islands, 18 Janu-
ary. Two males and two females were taken at both of the last two
localities. Two birds from the Paumotus are in rather worn plumage.
January specimens from Taritari and Rongelab have molted and are
in fresh plumage save for one bird (a female) from Rongelab. In it
the wings and tail show much wear. One male from the same locality
has a strong wash of rufous on the neck and upper breast. In one
male and one female from Taritari the dark markings of the throat and
upper breast are nearly obsolete.

18. Heteractitis incanus (GmeHn).

Scolopa.r iiKcinm Gmelin, Syst. nat., 1789, 1, pt. 2, p. 6.58. (Eimeo and
Palmerston Islands).

Nine specimens in the collection were taken at Nukuhi\a, Marquesas
Islands, 16 September; Rangiroa, 21 September, and Makemo, 20
October, Paumotu Islands; Funafuti, Ellice Islands, 24 December;
Tarawa, 3 January, and Taritari, 6 January, Gilbert Islands; and
Kusaie, Eastern Carolines, 9 February. A female taken at Rangiroa,
Paumotu Islands, 21 September, still retains a part of the barred adult
plumage on the under parts. The other specimens are in winter
plumage. This species seems to range in winter across most of Poly-
nesia and the bird secured at Kusaie marks a point near its westward
limit. In the present collection it was replaced at Uala by the allied
H. brevipes.

180 bulletin: museum of comparative zoology.

19. Heteractitis brevipes (Vieillot).

Totanus brevipes Vieillot, Nov. diet. hist, nat., 1816, 6, p. 410. (Timor).

One male was taken at Uala in the Middle Carolines, 16 February,
1900. Mathews (Birds of Australia, 1913, 3, pt. 3, p. 209) considers
Heteractitis brevipes a form of H. incanus, but the differences character-
izing it seem so constant as to establish it as a full species. Examina-
tion of a considerable series of tattlers in the U. S. N. M. collection
reveals no intergradation and all specimens examined could be deter-
mined as either H. incanus or H. brevipes at a glance. The differences
between the two have been well set forth by Dr. Stejneger (Bull. 29
U. S. N. M., 1885, p. 132).

20. Limosa lapponica baueri (Naumann).

Limosa baueri Naumann, Vogel Deutschl., 1836, 8, p. 429. (Australia).

A female collected on Funafuti 24 December, 1899, constitutes,
apparently, the first record of the bird in the Ellice Islands. Mathews
(Birds of Australia, 1913, 3, pt. 2, p. 191) has divided the genus Limosa,
as at present recognized, into two groups, proposing the name Vetola
for Limosa lapponica, a genus which, if recognized, must also include
Limosa hacmastica and L. fedoa. He restricts Limosa to the single
species Limosa limosa. and in his diagnosis gives the following as dis-
tinguishing Vetola from it: "the bill is proportionately shorter and
more slender and distinctly more upturned; the groove on the upper
mandible becomes obsolete at about three-quarters the length of the
culmen owing to the strong vertical compression of the upper mandible,
the groove on the lower mandible persists however as in Limosa. The
legs are short, the exposed tibia being less than the length of the middle
toe, the metatarsus is less than twice the middle toe and also less than
one-third the length of the wing, the scutellation of the front of the
metatarsus becomes irregular and broken up into hexagonal scales
towards the tibio-tarsal joint, whereas in Limosa the scutellation is
quite regular. The middle claw is normal, untoothed and short,
being one-fourth, or less, the length of the middle toe."

Examination of a series of specimens of the four species of godwits
included under the genus Limosa {sensu latu) fails to substantiate the

TOWNSEND AND WETMORE: THE BIRDS. 181

validity of all of the differences outlined by Mr. Mathews, which
apparently are based only on Limosa liinosa and L. lapponica. His
alleged differences will be taken up and considered in turn as he has
given them. The bill in the species limosa is longer than in haemastica
and lapponica, but shorter than in fedoa. In limosa the bill is more
slender and less robust than it is in fedoa, while it is nearly straight
in limosa, slightly upturned at the tip in lapponica, haemastica, and
fedoa. The groove on the upper mandible varies slightly in length
in all four species, but shows no specific differences in its development.
With regard to the legs limosa has the lower extremities relatively
slightly longer and somewhat more slender than in the other three
species under discussion. The tarsus in fedoa is as long as it is in
limosa, but the leg in the former is heavier and more bulky, and when
compared with the wing is relatively slightly shorter than it is in
limosa. The length of the crus when compared with the middle toe
without the claw may be more or less as it varies individually. The
measurement of the tarsus compared with that of the middle toe is
also a variable factor, and may be more or less than twice the middle
toe without the claw, depending upon the condition of the individual
specimen. The length of the tarsus varies also when compared with
the length of the wing according to the specimen in hand, and may
equal more or less than one third of the wing, irrespective of species.

The scutellation of the front of the tarsus shows no constant differ-
ence in the four species under consideration. In general the scutes
are transverse on the lower tarsus, and divided or broken toward the
tibia.

The distance that the broken scutes extend down the front of the
tarsus varies individually in all the species, and in addition there is a
tendency for single scutes on the front of the tarsus below the area to
become broken or divided. The development of the claw of the
middle toe is also a variable character in spite of what has been said
to the contrary. It is usually more slender in limosa than in the others,
and seems always to be slightly more elongate in that species; but
may be produced also and even pectinated in haemastica and fedoa,
while the outer margin is often thin and distinctly crenulated in lap-
ponica, with pectinate divisions indicated in some specimens.

To sum up the discussion as given above the valid differences
between these proposed genera are apparently as follows: —
a. Legs relatively slightly longer, more slender; toes relatively
slightly more slender; bill nearly straight, not distinctly up-
turned at distal end, somewhat flattened near tip Limosa.

182 bulletin: museum of comparative zoology.

aa. Legs relatively slightly shorter, somewhat heavier; toes rela-
tively slightly stronger; bill slightly upturned at distal end,
decidedly flattened for distal third (sometimes for slightly more).

VdoJa.
After careful consideration and reexamination of a series of the four
species these distinctions appear to be too slight and too inconstant
to merit recognition as of generic value. The genus Limosa is used
therefore in the present connection as the proper one for Limosa
lappo7iica haucri.

' 21. PisOBiA ACUMINATA (Horsfield).

Totanus acuminatus Horsfield, Trans. Linn. soc. London, 1821, 13, p. 192.
(Java) .

Three specimens were secured at Taritari in the Gilbert Islands on
6 January, 1900. The species does not appear to have been recorded
previously from this group.

22. Aechmorhynchus parvirostris (Peale).

Tringa parvirostris Peale, U. S. explor. exped., 1848, 8, p. 235. (Dog or
Honden, and Raraka Islands, Paumotii Group).

An adult female w^as shot on Whitsunday Atoll, Paumotu Islands,
31 October, 1899. Tringa parvirostris of Peale has been referred to
Tringa cancellata Gmelin (Syst. nat., 1789, 1, pt. 2, p. 675) which is
based on the Barred phalarope of Latham. In Latham's original
description (Gen. syn. birds, 1785, 3, pt. 1, p. 274) it is stated that the
bird described was in the collection of Sir Joseph Banks and that it
had come from Christmas Island. Latham remarks that the bill was
one inch long and that the under parts were white barred with dusky.
There are in the U. S. N. M. three of the five specimens collected by
Peale on Dog (or Honden) and Raraka Islands. On examining these
and the bird secured on Whitsunday Atoll it is found that in the
largest the bill is only 18 mm. long, and that in all the throat and
abdomen are plain and unmarked. The birds secured by Peale are
much stained and discolored, but in the fresh specimen taken by Dr.
Townsend the under parts have a distinct buffy tinge. It appears,
therefore, that Latham's Barred phalarope should be referred to
some other species than the present one. Dr. T. H. Streets, U. S. N.,

TOWNSEND AND WETMORE : THE BIRDS. 183

who made a collection of birds on Christmas Island in the Fanning
Group in 1874 (Bull. 7 U. S. N. M., 1877) did not record Aeckmo-
rhynchus jjarvirostris, and Christmas Island is two thousand miles
from the nearest point at which that species is known at present.

The female shot on Whitsunday Atoll is distinctly larger than the
specimens collected by Peale. The bill is longer, and the toes are
noticeable for their length. As Peale's specimens are more or less
stained and yellowed there can be no direct comparison in color and
as none of his specimens have the sex indicated on the labels it is proba-
ble that these differences in size may be merely sexual characters.

It is possible that AechmorhynchiLS parvirostris is an ancient species
now on the verge of extinction. The bird collected on Whitsunday
Atoll has three white edgings on the scapulars of the left side, and one
of Peale's specimens (U. S. N. M. 15,721) possesses an extra digit with
two phalanges and a small claw, growing from the base of the second
phalanx of the fourth toe on the left foot. These abnormalities may
mark degeneration due to inbreeding, or declining virility in the stock.

LARIDAE.
23. Angus stolidus pileatus (Scopoli).

Sterna pileata Scopoli, Deliciae florae et faunae Insubricae, 1786, pt. 2, p. 92.
(Philippines).

Four specimens, all adult males, were collected as follows: —
Kusaie, Eastern Caroline Islands, one, 8 February, 1900; Funafuti,
Ellice Islands, one, 24 December, 1899; Makemo, 22 October, 1899,
and Makatea, 6 October, 1899, both in the Paumotu Islands. Mathews
(Birds of Australia, 1912, 2, pt. 4, p. 411) assigns the name Anous s.
unicolor (Nordmann) to birds from the Society and Paumotu Islands
"and other South Pacific groups.". He remarks only that this pro-
posed form is larger than any of the others. xA.s it is found that birds
from the Straits of Malacca and the China Sea are fully as large, this
name is rejected and the specimens are placed under A. s. ■pileatus
(Scopoli). It may be remarked also that specimens of A. s. gala-
poycnsis Sharpe, a form distinguished by its darker coloration, have
the wing as long as birds from the Paumotu Islands. Measurements
of the specimens in the present collection are as follows : —

184 bulletin: museum of comparative zoology.

No. Sex Locality Wing Tail Culmen Tarsus

U.S.N.M. 212,148 cf Kusaie, East- 8 Feb.,

ern Carolines 1900 277.0 1.'3S.2 42.0 26.0
" 212,147 c^ Funafuti, El- 24 Dec,

lice Islands 1899 267.0 153.0 40.0 23.0

" 212,146 cf Makeino, Pau- 22 Oct.,

motu Islands 1899 284.0 170.0 41.. 5 25.0
" 212,145 d' Makatea, Pau- S Oct.,

motu Islands 1899 285.0 166.0 43.0 24.0

24. Megalopterus melanogenys (G. R. Gray).

Anous melanogenys Gray, Gen. birds, 1846, 3, p. 658, pi. 182. (No type-locality
assigned).

Three specimens come from the Paumotu Islands, a male taken at
Rangiroa, 21 September, 1899, a female at Makemo, 22 October, 1899,
and a third specimen that bears neither locality nor date. These
birds are all fully adult. From lack of sufficient material for com-
parison it is not practicable at present to designate to what subspecies
the birds from the Low Archipelago belong. Mathews (Nov. zool.,
1911, 18, p. 4) finds that Megalopterus Boie (Isis, 1826, p. 980) pre-
occupies Micranous Saunders (Bull. 23 Brit, ornith. club, 1895, p. 19).
Mathews (Birds of Australia, 1912, 2, pt. 4, p. 420) states further that
the name Aiioiis mitudus Boie (Isis, 1844, p. 188) "is applicable to
this species only, the bill ^characters being diagnostic." As Boie, in
describing the bill, remarks simply " Schnabel und Fiisse schwarz,
ersterer sehr schwach," his statement might apply to either the
present species or to M. tcnuiroslris. As the name cannot be fixed it
must be abandoned. The next available name is Anous melanogenys
G. R. Gray (Gen. birds, Jany., 1846, 3, p. 658). This preoccupies
the name Anous leucocapillus Gould (Proc. Zool. soc. London, 1845,
[Feby., 1846], p. 103) as this designation although in the volume of the
Proceedings of the Zoological Society for 1845 was not actually pub-
lished until February, 1846. The species therefore will stand as
Megalopterus melanogenys.

25. Procelsterna cerulea (F. D. Bennett).

Sterna cerulea Bennett, Narrative whaling voyage, 1S40, 2, p. 248. ("Christ-
mas Island and other low coral formations of the Pacific").

One specimen was taken, an adult female from Rangiroa, Paumotu

TOWNSEND AND WETMORE: THE BIRDS. 185

Islands, collected 22 September, 1899. From lack of material for
comparison it is not possible to assign a subspecific name to this bird.
According to Mathews (Birds of Australia, 1912, 2, pt. 4, p. 431),
specimens from the Paumotu, Marquesas, and Society Islands should
be called Procelsterna ccrulca teretirostris (Laf resnay e) .

26. Gygis alba pacifica (Lesson).

Sterna pacifica Lesson, Ann. sci. nat., 182.5, 4, p. 101. (Society Islands,
Paumotu Islands, and Bora Bora).

Three specimens were taken, a female at Tekokoto, 26 October,
1899, and two males at Akiaki, Paumotu Islands, 30 October, 1899.
Measurements of the specimens follow : —

No. Sex Locality Wing Tail Culmen Tarsus

M. C. Z. 81,929

(U.S. N.M. 212,1.52) cf Akiaki, Pau- 30 Oct.,

motu Islands 1899 249.0 124.0 40.2 1.3.8

'212,153 c? Akiaki, Pau- 30 Oct.,

motu Islands 1899 243.0 127.5 41.0 13.0

" 212,154 9 Tekokoto.Pau- 26 Oct.,

motu Islands 1899 243.0 117.0 40.0 13.5

27. Gygis microrhyncha Saunders.

Gygis microrhyncha Saunders, Proc. Zool. soc. London, 1876, p. 668. (Marque-
sas Group).

Four specimens were collected of which three, a male, a female, and
an alcoholic specimen whose sex is not known at present were taken
at Nukuhiva, Marquesas Islands, on 16 September, 1899. The fourth
is a mummy that probably should bear the same date. These birds
differ from the original description in having the shafts of the rectrices
blackish (not entirely white). In addition the second rectrix is the
longest instead of the third as is stated by Saunders. The black eye
ring is more distinct than in specimens of Gygis alba.

28. Sterna lunata Peale.

Sterna lunata Peale, U. S. explor. Exped., 1848, 8, p. 277. (Vincennes Island,
Paumotu Group).

One specimen, an immature bird, was collected at Rangiroa,
Paumotu Islands, 23 September, 1899.

186 bulletin: museum of comparative zoology.

29. Sterna sumatrana Raffles.

Sterna sumatrana Raffles, Trans. Linn. soc. London, 1821, 13, p. 329.

(Sumatra).

One specimen, a young bird not yet grown, in juvenal plumage was
taken at Arhno Atoll in the Marshall Islands, 26 January, 1900.
Mathews (Birds of Australia, 1912, 2, pt. 4, p. 372) gives two forms
under this species but seems uncertain as to the status of birds from
Polynesia as he says "birds from Fiji, Pelew Islands, and Phoenix
Island do not seem easily referable to the North-east Australian form."
The latter he has described as Sterna sumatrana kcmpi (Nov. zool.,
1912, 18, p. 210) with Torres Strait as the type-locality. The speci-
men in hand is too immature to permit satisfactory comparison.

30. Thalasseus bergii rectrirostris (Peale).

Sterna rectrirostris Peale, LT. S. explor. exped., 1848, 8, p. 281, pi. 75, fig. 2.
(Feejee Islands).

Three specimens were taken at Makemo in the Paumotu Islands, a
female 22 October, and two males 23 October, 1899. These birds are
typical of this subspecies and differ from T. b. pclccanoides (King) in
being slightly paler above. The differences noted are slight but seem
strong enough to uphold the separation of this form.

TRERONIDAE.

31. Ptilopus coralensis (Peale).

Ptilinopus coralensis Peale, U. S. explor. exped., 1848, 8, p. 190. (Carlshoff
or Aratica Island).

There are fourteen specimens of this fruit pigeon in the collection
taken in the Paumotu Islands at the following localities : — Makemo,
20 October; Niau, 7 October; Fakarava, 11 October and Makatea,
26 September and 6 October. The type of this species came from
Carlshoff or Aratica Island, and the specimens listed above give the
first definite records known for other islands, though Peale remarks
that the species was found on many of the low coral islets of the

TOWNSEND AND WETMORE: THE BIRDS. 187

Paumotu Group. The birds from Makatea have the hght edgings
of the secondaries paler than specimens from Niau, Fakarava, and
Makemo, but are otherwise the same. The type-specimen of Ptili-
nopus coralensis Peale was redescribed by Salvadori (Cat. birds Brit,
mus., 1893, 21, p. 105) as PtUopus smithsonianm, as he found that it
did not agree in color with the platfe as given by Peale. Peale's type
was a mounted bird originally, though now remade as a skin, and the
differences noted by Salvadori seem due to long exposure to dust and
light. PlUoims smithsonianus w^ill stand as a direct synonym of
P. coralensis with the same type-specimen extant for both (see Ogilvie-
Grant, Ibis, 1913, p. 349).

32. Ptilopus perousii (Peale).

Ptilinopus perousii Peale, U. S. explor. exped., 1848, 8, p. 195. (Upolu,
Samoan Islands).

Five specimens of this handsome pigeon were collected at Tongatabu
in the Tonga Islands on 30 November. One other has the locality
uncertain but probably came from this same island. Three males in a
series of four have an ochraceous orange band across the breast,
while in the fourth this band is barely indicated. One female has the
shorter under tail coverts tipped with red and their bases and the
longer feathers yellow. Another (place of capture somewhat uncer-
tain) has the under tail coverts entirely yellow.

33. Ptilopus dupetithouarsii (Neboux).

Cclumbia dupetithousarsii Neboux, Rev. zooL, 1840, p. 289. (Christina Island,
Marquesas Group).

Eight specimens, six of them males, were taken at Nukuhiva, Mar-
quesas Islands, on 16 September. The males vary in the amount of
reddish orange on the underparts. One adult female resembles the
males but has less of this bright color below than the more highly
colored specimens. An immature female has the reddish orange patch
below as in adults while the crown cap is grayer and much obscured
posteriorly by greenish tips on the feathers. The bill in this species
is dusky, the feet brownish.

188 bulletin: museum of comparative zoology.

34. Ptilopus porphyraceus (Temminck).

Columha porphyracea Temminck, Trans. Linn, soc, London, 1821, 13, p. 130.
(Tongatabu and Ulieta).

Eleven specimens were collected at the following localities : — Nine,
25 November; Eua, 28 November; Tongatabu, 30 November;
Nomuka, 2 December; and Vavau, 4 December, all in the Tonga
Islands; and Kambara, 7 December in the Fiji Islands. The species
does not seem to have been recorded from Kambara and Nomuka
before. The bird from the Fijis has been called Ptilopus porphyraceus
devientinae (Jacquinot and Pucheran) (Wiglesworth, Aves Polynesiae,
1891, p. 50). It is slightly paler on the breast and throat than the
average in the series but in this respect is equalled by one bird from
Eua. The under tail coverts are yellower also but other specimens
from the Tonga Islands resemble it closely. A bird in immature
plumage from Eua has the under tail coverts entirely yellow with no
orange at all, so that the depth and extent of the orange color in these
feathers seems to be dependent upon age. No females are represented
in the collection so that no comparison is possible between the two
sexes in regard to this character. It is worthy of note that two males,
apparently adult, from Nine have the under tail coverts entirely deep
yellow.

In an immature bird from Eua the aster-purple crown of the adult is
indicated on the forehead by a few new feathers at the .base of the cere.
Elsewhere the crown is green (between light hellebore and light elm-
green) like the back. All of the wing coverts, tertials, and scapulars
are tipped with yellow. The primaries are tipped with white, the
secondaries with white and margined with yellowish, and there are
obscure yellow tips on the feathers of the back forming slender cross-
bars. The terminal tail band is obsolete on the median pair of rec-
trices. The under tail coverts, abdomen, and a patch either side of
the rump are yellow, there is an obscure yellowish patch on the middle
of the abdomen and the feathers of the lower throat and breast are
tipped with yellow. There is no trace of a dark band on the lower
breast, and the concealed blue spots found in the adult near the tips
of tertials and scapulars are faintly indicated on some feathers in these
areas by obscure spots of slightly brighter green.

TOWNSEND AND WETMORE : THE BIRDS. 189

35. Ptilopus purpuratus (Gmelin).

Columba purpurata Gmelin, Syst. nat., 1789, 1, pt. 2, p. 7S4. (In insulis au-
stralis intra tropicos inclusis).

One specimen was collected at Tahiti in the Society Islands on
13 November, and two more were secured the following day. One of
these birds has a strong coppery reflection on the rectrices as in P.
coralensis. This color is arranged in narrow bands. The feet in all
three specimens are blackish.

36. Ptilopus hernsheimi Finsch.

Ptilopus hernsheimi Finsch, Journ. flir ornith., 1880, p. 303. (Kusaie, East-
ern Carolines).

A male, a female, two immature birds, and two mummies (formerly
preserved as alcoholic specimens) are in the collection from Kusaie,
8 and 9 February. In color the female is similar to the male save
that there is less of deep orange color on the longer under tail coverts.
The other sexual color differences indicated by Dr. Finsch in his
original description are not evident. The two immature birds differ
from the adults in having the crown-cap barely indicated on the fore-
head. In addition the feathers of back, rump, wing coverts, and under
parts are tipped lightly with pale yellow, the secondaries are tipped
with yellowish white, the primaries are pointed with white and the
terminal band on the rectrices is much narrower.

Measurements of the birds with sex indicated are as follows: —

No. Sex Wing Tail „ : Tarsus

Ciumc7i

U. S. N. M. 212,294 cT adult - 130.0 75.0 13.0 24.5

" 212,295 cf im. 125.5 67.0 14.0 23.0

" 212,293 9 adult 127.0 73.0 12,5 22.5

37. Ptilopus ponapensis (Finsch).

Ptilincpus ponapensis Finsch, Proc. Zool. soc. London, 1877, p. 779. (Ponape,
Caroline Islands).

Four males, one female, and another specimen, a mummy, with sex
not marked, were secured on the Island of Uala in the Middle Caro-

190 bulletin: museum of comparative zoology.

lines (Truk Group) on 16 Februar^^ These birds have a spot in
front of the eye that extends back as a narrow superciliary stripe,
and the middle of the throat and chin distinctly yellow, characters
that are not mentioned by Dr. Finsch in his original description, nor
by Count Salvadori in the British museum catalogue (1893, 21, p. 93).
A young male in juvenal plumage from Ponape has these same mark-
ings indicated, so that there are no differences evident in the speci-
mens at hand. Fully adult birds from Ponape are not available for
comparison. In these yellow markings P. ponapensis resembles
hcrnshcimi from the Eastern Carolines.

Measurements of the specimens with the sex indicated follow: —

No.

Sex

Wing

Tail

Exposed
Cuhnen

Tarsu

U. S. N.

.212,287 ■

130.0

74.0

15.0

22.5

212,288

133.5

73.0

13.5

24.0

. C. Z. 81,934 (

212,289)

1.37.0

76.0

14.0

24.0

. C. Z. 81,935 (

212,290)

131.5

73 . 5

14.0

24.0

212,291

126.0

68.5

14.5

22.0

38. Globicera AURORAS (Peale).

Carpophaga aurorae Peale, U. S. explor. exped., 1848, 8, p. 201. (Aurora or
Makatea Island).

A female was collected on Makatea Island 6 October, 1899. This
bird is molting, and has not yet cast the six inner pairs of second-
aries nor the second pair of rectrices. These old feathers are dull
brown in color with a very slight gloss of blue, so that they are much
duller than the bright new feathers. From a study of skins (including
the types) in the U. S. N. M. collection it appears that Glohiccra
wilkesii (Peale) (U. S. explor. exped., 1848, 8, p. 203) described from
Tahiti is a synonj'm of Globicera aurorae, as the differences between
these supposed species are merely those of age (and perhaps of sex).
G. wilkesii was said to be darker in color than G. aurorae and to have a
smaller knob on the cere. In addition Peale ascribed to it twelve
rectrices instead of fourteen but this difference in the number of tail
feathers is due to an imperfection in the specimen that he chose for
type. As regards color, specimens from Makatea are found that are
as dark as those from Tahiti in corresponding plumage and the lighter
birds may occur in either locality. The size and shape of the cere

TOWNSEND AND WETMORE: THE BIRDS. 191

varies in individuals, regardless of locality, and may be seasonal in its
growth, or may differ in the sexes. No differences in measurements
are apparent, but only a few of the specimens examined have the sex
given, and in part of those there is some doubt as to the correctness
of the determination. There is apparently only one fully adult in the
series of nine available, and this bird is sexed as a male. Peale
remarks (Opuscit., p. 204) that a specimen of G. wilkcsri was obtained
by Captain Wilkes on Aurora (Makatea) Island, and this specimen
is still in the U. S. N. M.

39. Globicera pacifica (Gmelin).

Columba pacifica Gmelin, Syst. nat., 1789, 1, pt. 2, p. 777. (Friendly Islands).

A male and a female were secured on Funafuti in the Ellice Islands,
25 December, 1899. A species of pigeon was reported from this
atoll for many years, and the fact that it belonged to this form was
finallv established by Mr. A. J. North (Rec. iVustr. mus., 1898, 3,
p. 85).

40. Globicera oceanic a oceanica (Desmarest).

Columba oceanica Desmarest, Diet. sci. nat., 1826, 40, p. 316. (Ualan -
Kusaie).

An immature male was taken 8 February, and a female, 9 Febru-
ary, at Kusaie, the type-locality. The male is apparently fully grown
but shows differences in color from the female that seem to be due to
immaturity. The throat, malar region, and space behind the eye
are more extensively white, and the feathers of the breast have faintly
indicated paler tips. The lower breast is paler than the upper breast
and the rufous of the lower breast is restricted by this paler color.

Measurements of these two specimens are as follows: —

Uulmen

No.

Sex

Wing

Tail

frcm Cere

Tarsus

U.S. N. M. 212,242

9 adult

216.0

1.39.0

17.0

28 . 5

212,24.3

c^ im.

208.0

128.0

17.5

31.0

41. Globicera oceanica townsendi, subsp. nov.

Characters. — Similar to Globicera oceanica oceanica (Desmarest)
from Kusaie, Caroline Islands, but darker on upper breast, foreneck.

192 bulletin: museum of comparative zoology.

hindneck, and upper back; no whitish Hne indicated on lower eyelid
beneath eye; and under tail coverts paler.

Type.— \J. S. N. M. 212,240. Adult female. Polynesia: Eastern
Caroline Islands; Ponape, 12 February, 1900. H. F. Moore.

Description. — Forehead, at base of bill white; crown, nape, hind-
neck, and upper scapular region between deep and dark neutral gray,
changing to neutral gray immediately behind the white on the fore-
head; wing coverts, scapulars, back, rump, and upper tail coverts
dull blackish green, the feathers iridescent, with darker green or very
dark blue more or less distinctly indicated at the tips; primaries dull
greenish black no. 2; visible portions of rectrices dull blackish green,
the central pair with faintly indicated narrow darker crossbars;
malar region, chin, and upper throat, whitish; side of head, throat,
and sides of neck between deep and dark olive-gray; breast between
mouse-gray and deep olive-grg,y; abdomen, tibiae, and under tail
coverts russet; sides, flanks, and axillars dark neutral gray; bend
of wing washed with rufous; bill and cere black; tarsus and toes
brownish yellow, nails black (from dried skin).

Measurements. — Females (two specimens, no males seen) wing
221-223; tail 142.5-147; culmen (from cere) 17.5-18; tarsus 32-33.5.

Range. — Island of Ponape.

Remarks. — Two adult females of this fine pigeon were collected on
Ponape, 11, 12 February, 1900. Both agree closely in color but one is
slightly larger than the other. These two are distinctly darker than
specimens from Kusaie and may be distinguished from them without
difficulty. The type has the bend of the wing washed with rufous
but this color is lacking in the second specimen.

LORIIDAE.

42. Eos RUBiGiNOSA (Bonaparte).

Chalcopsitta ruhiginosa Bonaparte, Conspec. gen. avium, 1850, 1, p. 3. ("ex
insulis Barbay, et Guebe." Later attributed correctly to Ponape.
Cf. von Pelzeln, Novara exped. Vogel, 1865, p. 99).

Two males, two females, and two mummies (dried from alcohol)
were collected at Ponape in the Eastern Caroline Islands, 12 Febru-
ary. Males and females are alike in coloration and size. This parrot
was supposed to be a native of Waigiou for many years until discovered
on Ponape ("Puynipet") by the Novara expedition.

TOWNSEND AND WETMORE: THE BIRDS. 193

43. ViNi AUSTRALis (Gmelin).

Psittacus australis Gmelin, Syst. nat., 1788, 1, pt. 1, p. 329. (Samoan Islands).

Three birds were secured at Niue or Savage Island, 25 November,
and four were taken at Upolu in the Samoan Islands. The birds
from the two localities appear identical in color and size. It is proba-
ble that these small parrots have been carried from island to island
as cage-birds.

44. Choriphilus peruvianus (Miiller).

Psittacus peruvianus Miiller, Natursyst. Suppl., 1776, p. 80. ("Peru."
Based on Buff on. Type-locality is here given as Tahiti, cf. Daubenton,
Planches enlum., no. 455, fig. 2).

There are in the collection four skins of this small parrot from
Rangiroa, Paumotu Islands, taken 21, 22, and 23 September, seven
from Bora Bora, Society Islands, collected 17 November, and seven
from Aitutaki, Cook Islands, secured 21 November. In addition
five birds from Bora Bora were preserved as alcoholic specimens.
The name Psittacus taitianus (Gmelin) (Syst. nat., 1788. 1, pt. 1,
p. 329) has been commonly applied to this parrot with Psittacus
peruvianus given as a synonym. As Mtiller's name antedates that
used by Gmelin and as there is no ground for supposing that his bird
is not this species Psittacus peruvianus must be accepted. Miiller
gives Peru as the type-locality which is obviously erroneous. The
type-locality is hereby restricted to Tahiti.

Two male birds from Rangiroa are in immature plumage. In both
the breast is blackish. In one a few white feathers show on the ear
coverts and throat while in the Qther the cheeks and a spot on the
breast are white. The dark immature bird from Tahiti has been
described by Sparrman (Mus. Carlson., 1787, fasc. 2, no. 27, pi. 27) as
Psittacus cyaficiis. More recently Mr. Scott B. Wilson (Ibis, 1907,
p. 379, pi. 8) has named a bird in this same dark plumage from Bora
Bora in the Society Group, calling it Coriphilus cyannis. Later (Ibis,
1907, p. 653) on learning of the previous use of Psitticus cyaneus by
Sparrman, Wilson renamed his bird Coriphilus cyanescens. On refer-
ring to his original description and the colored plate accompanying it
there can be no question that this is an immature specimen of C. peru-
vianus.

194 bulletin: museum of comparative zoology.

The species does not seem to have been recorded from the Cook
Islands before and though it is known from the Paumotu Islands
apparently none have been collected previously on Rangiroa. Careful
comparison of the series from the three localities represented reveals
no differences in coloration but there are some slight diiTerences
evident in size. Two males from Aitutake average smaller than
males from Bora Bora, Society Islands. Males from Rangiroa are
likewise smaller than those from Bora Bora. From the present series
however, it does not appear that these differences merit distinction
by name.

Average measurements are as follows (in millimeters) : —

Sex

Locality

Wing

Tail

Cvlmen
from Cere

Tarsm

2 d^cf

Aitutake

107.5

68.2

9.5

14.1

7 cj^c?

Bora Bora

114.6

70.5

10.1

15.8

4 cT^cf Rangiroa 109.7 68.3 10.2 14.3

. Females are represented from Aitutaki only. Four have the average
wing measurement 104.9, tail 65.4, culmen from cere 9.4 and tarsus
14.6.

CUCULIDAE.
45. Urodynamis taitensis taitensis (Sparrman).

Cuculus taitensis Sparrman, Mus. Carlson., fasc. 2, 1787, 32, pi. 32. (Tahiti).

An adult (sex not determined) was shot on Funafuti in the EUice
Group, 24 December. This cuckoo is said to be resident in both
the Ellice and Gilbert Islands. In the latter group there are no other
land birds, and in the Ellice Islands a pigeon is the only land bird
known other than the cuckoo, so that there has been considerable
speculation as to in what way this species practiced its parasitic habit
of foisting its eggs upon other species for incubation and the rearing
of its young. Concerning this the following observations by Mr.
Swayne are of interest: — "In August last year I was at the Island of
Niu, in the Ellice Group, and while walking through the island along
with the local trader we passed a clump of ' buka ' trees, in which, as
is common throughout the Ellice Islands, there were numbers of the
Noddies {Anous , stolid us) nesting. I noticed that in one tree the birds
were much disturbed and apparently frightened. The trader explained

TOWNSEND AND WETMORE : THE BIRDS. 195

that the birds were disturbed by a 'Hawk.' We remained some time
watching, and I saw our friend the Cuckoo drive a Noddy out of the
nest and take possession of it, while the old birds and apparent pro-
prietors tried in vain to dislodge the intruder. * * *

" I do not doubt that the Cuckoo was about to lay. * * *
"Although I offered rewards to the natives on many .islands, I
never was able to get an egg of the Cuckoo. In the Gilberts the people
say they have never seen eggs or young, and, as I told you, they hold
the tradition that the female takes a portion of the covering of the
young palm-leaf and flying up with it deposits it on a cloud, lays her
egg on it, where it is hatched by the sun." North, Proc. Zool. soc.
London, 1896, p. 934.

It would be strange indeed if this remarkable bird could survive
under the care and feeding of the Noddy, but this seems more plaus-
ible than that its young should be able to thrive (in the EUice Islands)
upon the vegetable food given young pigeons for their sustenance.
Certainlv the former belief is the more credible.

ALCEDINIDAE.

46. Sauropatis mediocris (Sharpe).

Halcyon mediocris Sharpe, Cat birds Brit, mus., 1892, 17, p. 260. (Ponape).

A female was taken on Ponape' in the Eastern Caroline Islands,
12 February, 1900. In his original description Sharpe designates
this bird as subspecies b of S. cinnamomina (Swainson) (though he
uses a binomial name for it). The bird at hand differs from females
of S. cmnavtomina in having the under parts entirely white, the crown
paler and the collar on the hind neck white instead of cinnamon. In
addition the white collar is bordered behind by black and the back
is more bluish. The differences are so great that Sauropatis mediocris
seems (from the present material at least) to represent a full species.^
S. lucdiocris agrees with the following species S. sacra in the form of its
bill, and in ha\'ing the tenth (outermost) primary shorter than the
sixth and longer than the fifth.

'As has been pointed out by Seale (Occas. papers Bernice Pau ah i Bishop mus., 1901, 1,
no. 3, p. 40) Halcyon rujigularis Sharpe (Cat. birds Brit, mus., 1892. 17, p. 260) based on a skin
received from the Zoological society of London, is the female of Sauropatis cinnamomina
(Swainson),

196 bulletin: museum of comparative zoology.

47. Sauropatis sacra sacra (Gmelin).

Akedo sacra Gmelin, Syst. nat., 1788, 1, pt. 1, p. 453. (Type-locality hereby
restricted to Tongatabu Island).

One male was taken on the Island of Tongatabu, 30 November.
There are two other birds, a male and a female from the island of
Nomuka taken 2 December, that have been referred to this form but
are not typical of it as they are slightly duller in color than the bird
from Tongatabu, and in addition are considerably smaller. Meas-
urements are as follows: —

No.

Sex

Locality

Wing

Tail

Culmen
from Base

U. S. N. M. 212,341

Tongatabu Id.

105.0

73.5

44.0

212,340

Nomuka Island

97.0

68.5

43.0

" 212,339

u u

101.0

72.5

43.5

The birds from Nomuka are in worn plumage, which might account
in part for the shortness of wing and tail. The single bird from
Tongatabu, however, has a large heavy bill that is noticeably stronger
and broader at the base than in any other specimen in the various
forms of this species examined. Females of Sauropatis sacra in general
differ from males in having the superciliary stripe whiter, in being
decidedly more greenish above, and in having little or no blue appar-
ent in the blackish feathers on the anterior surface of the tibia. In
addition females are the larger of the two sexes.

In treating the geographical forms of Sauropatis sacra it becomes
necessary to restrict the typical subspecies. Therefore, I designate
the island of Tongatabu, in the Tonga Islands, as the type-locality
of Sauropatis sacra sacra. The Alcedo sacra of Gmelin was based
upon Latham's sacred Kingfisher (Latham, Gen. syn. birds, 1782, 1,
pt. 2, p. 621). Latham states that his bird had a blue band on the
hind neck, in this agreeing with the forms found in the Tonga Islands.
x\n island in this group therefore is selected as the type-locality for
the typical form, as birds from the Fiji Islands have this band black
or with only a trace of blue. The latter are to be known as Sauro-
patis sacra mtiensis (Peale) (U. S. explor. exped., 1848, 8, p. 156).

This species has the tenth (outermost) primary- slightly shorter
than the sixth and longer than the fifth, and so does not agree wholly
with the figure of the wing of Sauropatis sanctus given by Mathews

TOWNSEND AND WETMORE : THE BIRDS. 197

(Austr. avian record, 1912, 1, p. 107) in recognizing the genus Sauro-
patis of Cabanis and Heine. In that species the tenth primary is
longer than the seventh and in Halcyon senegalensis, which is also
figured by Mathews, the tenth primary is considerably shorter than
the fifth. In the form of its bill Sauropatis sacra agrees closely with
»S. sanctm.

48. Sauropatis sacra rabulata, subsp. nov.

Characters. — Similar to Sauropatis sacra sacra (Gmelin) but darker,
less greenish above, especially on head and rump.

Te/pe.— U. S. N. M. 212,343. Male. Polynesia: Tonga Islands;
Eua, 28 November, 1899. C. H. Townsend.

Description. — Crown and collar on hind neck dusky greenish blue;
back, scapulars, tertials, and inner secondaries chessylite-blue; rump
near motmot-blue; lesser and middle wing coverts Blanc's blue, the
feathers edged more or less with Mathews's blue; first primary, tips
and inner webs of other primaries, and under side of rectrices dull
black; greater wing coverts, outer webs of primaries (except the
first) and upper side of rectrices dusky greenish blue; superciliary
stripe white mixed with cinnamon-bufF; behind the eye this stripe is
entirely cinnamon-buff, is broadened and unites with its fellow from
the opposite side ; malar stripe, extending from gape to unite with blue
stripe on hind neck, chessylite-blue, the feathers black underneath
so that the two colors are mixed; lores and a narrow line under eye
black; a very narrow line of chessylite-blue over eye; spot on lower
eyelid, collar on hind neck, and entire under parts white; anterior side
of tibia blackish, with capri-blue tips on a few feathers.

Measurements. — Males (two specimens), wing 100-101; tail 69-70;
culmen from base 41^2.5; tarsus^ 15-15.3.

Female (one specimen) wing 104. tail 73.5, culmen from base 44,
tarsus 16.5.

Range. — Island of Eua, Tonga Group.

Remarks. — The superciliary stripe in fully adult birds is white
(as is shown by a female specimen) so that the type is hot quite in
full plumage. Adults and young after their first molt seem to differ
in no o+her way.

The Alcedo sacra of Gmelin (Syst. nat., 1788, i, pt. 1. p. 453) is
based upon the Sacred Kingfisher of Latham (Gen. syn. birds, 1782,
1, pt. 2, p. 621) who described the bird from the Leverian Museum.

198 bulletin: museum of comparative zoology.

In his original description Latham states that his bird has " under the
blue beneath the eye, a narrow orange ferruginous stripe." This
marking is not found in the present species; and Von Pelzeln (Ibis,
1873, p. 19) who has examined Latham's type says that this marking
is not indicated save for a slight tinge of yellowish under the auricular
region. Latham states that his Sacred kingfisher inhabited Otaheite
and the other Society Islands, an obvious error, as the species is not
known save from the Tonga and Fiji groups.

Four specimens (three skins and one bird in alcohol) upon which
this form is based were taken at Eua, Tonga Islands, on 28 November.

49. Sauropatis sacra celada, subsp. nov.

Characters. — Similar to Sauropatis sacra sacra (Gmelin) but lighter,
more greenish above, especially on the crown; malar stripe with
little or no black indicated save at its posterior margin.

Type — V.S.N. M. 212,M7. Male. Polynesia: Tonga Islands;
Vavau, 4 December, 1899. C. H. Townsend.

Description. — Crown, band on hind neck, back, scapulars, and
tertials capri-blue; upper tail coverts bremen-blue; first primary,
tips and inner webs of other primaries, and under side of rectrices
blackish; outer webs of primaries (save first), greater wing coverts,
and upper side of rectrices dark chessylite-blue ; lesser and middle
wing coverts china-blue; superciliary stripe white, mixed more or
less with ochraceous buff, the stripes from either side meeting on back
of head; lores black mixed with white, the black glossed with blue;
malar stripe Blanc's blue; spot on lower eyelid, and entire under parts
white; anterior side of tibia Blanc's blue, the feathers blackish basally.

Measurements. — Males (three specimens), wing 101-102.5 (101.8);
tail 70.5-73 (72.0); culnien from base 41-42 (41.6); tarsus 16-16.5
(16.1).

Female (one specimen), wing 106.0, tail 76.5, culmen from base 45,
tarsus 17.

Range. — Island of Vavau, Tonga group.

Remarks. — Four specimens of this form were taken at Vavau,
Tonga Islands, 4 December. One bird, a male, nearly adult, is
somewhat darker than the other two but is distinguishable from S. s.
sacra. All of the males have more or less of buffy in the superciliary
stripe a character which as has been noted above seems to be an indi-
cation of immaturity in this species, though this buffy color is less in

TOWNSEND AND WETMORE : THE BIRDS. 199

amount in females than in males. The female specimen has no
locality indicated on its label, but from its coloration is identified as
belonging without doubt to this form.

50. Halcyon sacra vitiensis (Peale).

Dacelo vitiensis Peale, U. S. explor. exped., 1848, 8, p. 156. (Venua-levu, Feejee
Islands) .

One female referred to this form was taken at Kambara in the Fiji
Islands, 7 December. Birds examined from the Fiji Islands differ
constantly from specimens in the Tonga Group in having the dark
band on the hind neck black, or with only an admixture of blue. In
Tonga Island birds this band is entirely blue. Fijian birds too appear
to be constantly smaller. Measurements of the bird taken on Kam-
bara are as follows : —

,, „ TTT. m -I Culmen from

No. Sex Wing Tail „ •'

Base

U. S. N. M. 212,-338, 9 93.0 65.5 42.5

51. ToDiRHAMPHUS RECURViROSTRis Lafrcsnayc.

Todiramphus recurvirostris Lafresnaye, Rev. zool., 1842, p. 134. ("in insulis
Maris Australis").

A single bird (sex not marked) was taken at Upolu in the Samoan
Islands.

52. ToDiRHAMPHUS TUTUS (Gmelin).

Alcedo tuta Gmelin, Syst. nat., 1788, 1, pt. 1, p. 453. (Tahiti).

Five specimens were collected at Bora Bora in the Society Islands,
17 November. An immature male has the feathers of the upper
breast buffy, with blackish cross bars forming a dark band across the
breast. Three of the birds taken have the forehead white, while in
the other two it is the same color as the crown, with white borders on
the feathers. Immature birds are more greenish above than the
adults.

Sharpe has stated (Hist, collections Brit. mus. Birds, 1906, p. 182)
that the "Respected" and "Venerated" kingfishers described by

200 bulletin: museum of comparative zoology.

Latham (Gen. syn. birds, 1782, 1, pt. 2, p. 623, 624) upon which
Gmehn (Syst. nat., 1788, 1, pt. 1, p. 453) founded his species Alcedo
tuta and A. venerata, "seem to be the same species, and hence Todi-
rhamyhiis tutus, Sharpe (nee. Gm.), Cat. birds, 17, p. 291, will require
another name, which I propose should be Todirhamphus wiglcsworthi,
in memory of the young explorer who did such good work as the his-
torian of the Pacific Avifauna." On turning to Latham's descriptions
it is found that the diagnosis of the "Venerated" Kingfisher is readily
applicable to specimens of Todirhamphus mneratus at hand. His
note on a band of glossy green on the hind neck "at which place it
inclines to white" is true, as in some individuals white markings
occur on the feathers in this region. In his account of the "Re-
spected" Kingfisher Latham says distinctly "over the eye a white
streak" which is one of the prominent differences between T. tutus
and T. veneratus. So that there is no question but the "Respected"
and the "Venerated" Kingfishers of Latham refer to separate species.
From this it appears that the name Alcedo tuta of Gmelin is valid
and that Todirhamphus wiglesworthi Sharpe must be placed in the
synonymy of this species.

53. Todirhamphus veneratus (Gmelin).

Alcedo venerata Gmelin, Syst. nat., 1788, 1, pt. 1, p. 4.53. (Said by Latham to
come from Apia. As the bird does not occur there the typ34ocality is
hereby stated to be Tahiti).

Five males and an immature female were secured at Tahiti, 3
October and 14 November, 1899. The immature bird has a broad
brown band across the upper breast, and is brown with only a tinge
of green above. The males all show a slight amount of brown on
either side of the breast and in one a broken band is indicated by
slender shaft streaks on the feathers of the upper breast.

MICROPODIDAE.

54. Collocalia francica townsendi Oberholser,

Collocalia francica townsendi Oberholser, Proc. Acad. nat. sci. Philad., 1906,
p. 197. (Eua, Tonga Islands).

Three specimens of this swift were secured in the Tonga Islands, one
at Nine, 25 November, one (the type) at Eua 28 November, and one

TOWNSEND AND WETMORE : THE BIRDS. 201

at Vavau, 4 December. The bird from Niue has the tips of the breast
feathers entirely worn away.

55. CoLLOCALiA THESPESiA Oberholser.

Collocalia thespesia Oberholser, Proc. Acad. nat. sci. Philad., 1906, p. 195
(Tahiti).

The type of this species, a female, was collected at Tahiti, Society
Islands, 14 November, 1899.

56. Collocalia ocista Oberholser.

CoUocalia ocista Oberholser, Proc. Acad. nat. Sci. Philad., p. 184. (Nuku-
hiva) .

Two specimens (including the type) were secured at Nukuhiva,
16 September, and one at Tahiti, 13 November. The bird from
Tahiti, a male, has the breast feathers much worn.

HIRUNDINIDAE.

57. Hypurolepis tahitica (Gmelin).

Hirundo tahitica Gmelin, Syst. nat., 1789, 1, p. 2, p. 1016. (Tahiti).

Three were taken on Nomuka, Tonga Islands, 2 December.
Only one of these, an adult female, has the sex indicated without
question. The remaining two are immature birds that differ from the
adult in having the chestnut of the forehead faintly indicated anteriorly
and obsolete behind. In addition the blue-black spots on the under
tail coverts are lacking, and in one bird there are faintly indicated pale
margins on the feathers of the back. This species is known from the
Tonga Islands but has not been reported from Nomuka previously.
The large, broadened bill, characteristic of the genus Hypurolepis
Gould, reaches its maximum development in this species, and appears
remarkably strong and heav'y for a swallow.

MUSCICAPIDAE.

Haplornis, nom. nov.

The name Muscylva Lesson was first proposed in vernacular form
in Lesson's Traite d'ornith., 1830; p. 385, with a proper diagnosis. In

202 bulletin: museum of comparative zoology.

a following part of this same work on page 656 (published in 1831)
the name Muscylva occurs in italics, this typography indicating that
the author used it as a technical or Latin name, his intention in regard
to this usage being outlined on page 651. Muscylva is therefore to
be quoted from this second reference. Seven species are included in
it by Lesson (p. 386) as follows :

Mvscicapa leiicogaster, Mus. de Paris. Cayenne. (Poiteau).

Muscicapa rvfiventer, Mus. de Paris. Nouvelle-Hollande (Peron).

Muscicapa alhogularis Mus. de Paris. Bengale. (Mace).

Muscicapa aurocapiUus, Mus. de Paris.

Muscicapa caendca, Gm.; le Petit azur; Enl., 666, fig. 1. Des
lies Philippines, du Bengale.

Muscicapa luzoniensis Gm.; Levaill., pi. 151, fig. 1. De Mada-
gascar.

Muscicapa rufiventer Gm.; I'Oranor, Levaill., pi. 155, fig. 1.
De Batavia. (Diard.)

The first four of these names as listed above are nomina nuda at this
place, although Pucheran (Arch. Mus. hist, nat., 1855, 7, p. 333) found
that the first is equi^•alent to Rhipidura pcctoralis (Jerdon) and the
second to Rhipidura rufivenlris (Vieillot). The third was described
later by Lesson (Belanger's Voyage In des Orientales^ 1831, pi. 264) as
Muscicapa {Muscylva) albogularis, which equals Rhipidura cdbicoUis
(Vieillot). Pucheran states that he was unable to find the type of
Muscicapa auricapillus.

On examining the remaining species in turn it is found that Musci-
capa caerulea Gmelin is now placed in the genus H^^pothymis, while
Muscicapa luzoniensis Gmelin is in the genus Penthornis. The
seventh species, Muscicapa rufiventris Gmelin, is apparently still
unidentified, although "I'Oranor" of Levaillant is considered to be
Pericrocotiis pcrcgrinus (Linne). From this consideration it appears
that of the seven names mentioned by Lesson under the genus Mus-
cylva, only the last three are at all recognizable at the place of original
publication, as the first four are nomina nuda. From this it appears
therefore that G. R. Gray, (Handlist of birds, 1869, 1, p. 349) was in
error when he restricted Muscylva to the first species given by Lesson,
Muscicapa leucogastcr. On the page cited he adopted Muscylva as a
subgenus of Todirostrum, and under it listed M. Icucogaster "Less,
ex Mus. Par." from " Cayenne." On page 332 of the same volume of
this publication he had already included Muscylva " Homb. & J."
as a subgenus of Rhipidura.

TOWNSEND AND WETMORE". THE BIRDS. 203

The type of Muscylva Lesson, 1831, has never been properly desig-
nated, although Lesson (Compl. oeu\Tes Buffon, 1837, 8, p. 366) him-
self cited " MnscicajM aJhogidaris" as the type. As it has been shown
that this species was a nomen. miduvi in the original reference this
designation is not valid. To eliminate the name Muscylva Lesson
1831 the type is here fixed as Muscicapa caerulea Gmelin, reducing
Muscylva to a synonym of Hyyotkymis Boie, 1826.

In 1846, G. R. Gray (Gen. birds, 1846, 1, p. 258) named a bird from
the Fiji Islands Rhipidura lessoni, basing the species on the " Muscylva
de Lesson" of Hombron and Jacquinot (Voy. Pole Sud. Zool., October,
1844, 3, pi. 11, fig. 2, Oiseaux). This bird was later described by
Jacquinot and Pucheran on page 75, vol. 3, of the text of the work just
cited, published in 1853, as Muscylva lessoni. With it was described
another species, M. pccforalis. Muscylva was not designated as a
new genus by these authors and it was not intended as new, but the
name has been generally accredited to them since the date of its
appearance in the Catalogue of birds in the British Museum (Sharpe,
1879, 4, p. 233). It is evident from the outline above that the status
of Muscylva as a generic term was in hopeless confusion, so that it
has seemed best to eliminate it and substitute a new name to avoid
future complications. Following are emendations of Muscylva that
have appeared:

Mvscicylva Gray, Gen. birds, 1849, 3, app., p. 53.

Muscisyhia Agassiz, Nom. zool. Aves, 1841, p. 88. (Not

Muscisyhia Hodgson, 1844 and 1845).
Miiscyla Gray, Cat. gen. and subgen. birds, 1855, p. 51.

58. Haplornis lessoni (G. R. Gray).

Rhipidura lessoni Gray, Gen. birds, 1846, p. 258. (Fiji Islands).

There are in the collection four "mummies" of this species, with
sex not indicated, that were collected at Viti Levu, Fiji Islands,
16 December. The specimen of this bird that Peale (U. S. explor.
exped., 1848, 8, p. 101) described as Monarcha cinereus was secured
on this same island.

59. Metabolus rugensis (Hombron and Jacquinot).

Muscicapa rugensis Hombron and Jacquinot, Ann. sci. nat., 1841, ser. 2, 16,
p. 312. (Ruk Islands).

An adult male was secured at Uala in the Truk Group (Middle

204 bulletin: museum of comparative zoology.

Caroline Islands). 16 February, 1900. This bird has extensive dark
markings on the inner webs of the third, fourth, fifth, and sixth
primaries and smaller areas on the other primaries, save the first,
which is entirely white with a dark shaft. A spot in the center of the
forehead (entirely surrounded by black) is white and there are two
black feathers on the right side of the hind neck.

60. Rhipidura kubaryi Finsch.

Rhipidura kubaryi Finsch, Proc. Zool. soc. London, 187.5, P- 644. (Ponape).

A male secured on Ponape in the Eastern Carolines on 12 February,
1900, has the following measurements: — wing 76, tail 89.5, exposed
culmen 10.5, tarsus 20.5.

61. Myiagra PLUTO Finsch.

Myiagra pluto Finsch, Proc. Zool. soc. London, 1875, p. 644. (Ponape).

Two males, a female, and one other bird (a mummy, unsexed) were
collected at Ponape, 11 February, 1900. Both males have a brown-
ish wash on the feathers of the chin, throat, and upper breast, a char-
acter that Dr. Finsch in his original description assigned to the female.
This is indicated only slightly in the female in the present collection.
The crown cap in both sexes is distinctly darker than the back with a
sharp line of demarcation behind. It is probable that the two males
are immature. The mummy may be an almost adult male, as it has
the wash barely indicated on chin and throat, and in addition has
the feathers of the upper breast glossed with bluish.

Measurements of the birds with sex indicated are as follows : — •

No.

Sex

Wijig

Tail

Exposed
Culmen

Tarsus

U. S. N. M. 212,467

80.0

74.. 3

12.5

20.5

212,468

•78.51

76.0

12.5

20.0

" 212,470

81. 0

75.0

14.0

22.0

62. Myiagra oceanica Jacquinot and Pucheran.

Myiagra oceanica Jacquinot and Pucheran, Voy. Pole Sud. Zool., 1853, 3,
p. 77. (Hogoleu).

A male and a female were secured at Uala in the Truk Group,

' Wing somewhat worn/

TOWNSEND AND WETMOKE : THE BIRDS. 205

Middle Carolines, 16 February. The measurements of these two
birds are as follows:

No.

Sex

Wing

Tail

Exposed
Cidmen

Tarsu

U. S. N. M. 212,471

80.5

67,8

15,0

20,0

212,472

76,5

6.3.5

15.0

21,5

63. Myiagra townsendi, sp. nov.

Characters. — Similar to Myiagra vanikorenm^ (Quoy and Gaimard)
but adult male with back heavily glossed with green, glossy green of
anterior under parts covering entire upper breast, posterior under
parts darker, thighs blackish, under wing coverts darker, bill larger,
tail and tarsus longer: Female with outer web of outer tail feathers
extensively paler, crown grayer, back strongly washed with brown,
and loral region distinctly lighter than feathers of crown.

Typc.— V. S. N. M. 212,464. Adult male. Polynesia: Fiji
Islands; Kambara, 7 December, 1899, C. H. Townsend.

Description. — Crown, nape, upper back, sides of head, throat, and
upper breast iridescent greenish slate-black; lower back, rump, and
upper tail coverts between slate-gray and slate color, the feathers
glossed with deep slate-green; wing and tail feathers sooty black;
lesser and middle coverts iridescent greenish slate-black; greater
coverts black with outer webs of feathers glossed with iridescent
greenish slate-black; inner tertials more or less iridescent; lower
breast amber-brown, becoming ochraceous tawny on sides and flanks,
and merging into warm buff on the middle of the abdomen, lower tail
coverts and sides of rump; thighs black, the feathers tipped with
brown; under wing coverts light buff tipped with warm buff; bill and
tarsus (in dried skin) black.

Measurements. — Males (two specimens) wing 76.0-77.5; tail 67.0;
exposed culmen 14.5-15.2; tarsus 18.5-19.0.

Range. — Kambara, Fiji Islands.

Remarks. — Four specimens were collected on Kambara in the Fiji
Islands, 7 December. ,Two (including the Type) are adult males
and a third with the sex not indicated is supposed to be a female. A
description of this latter bird is as follows : — Crown and hind neck
dusky green-gray; sides of head and collar on hind neck slate color;
anterior portion of crown washed with mouse-gray; lores indistinctly
whitish; back, scapulars, rump, and upper tail coverts between buffy

206 bulletin: museum of comparative zoology.

brown and olive-brown; wings and tail dull black; lesser wing coverts
between buffy brown and olive-brown; middle and greater coverts
dull black tipped with this same brown; inner secondaries margined
all around with light buff; rectrices tipped indistinctly with whitish;
outer web of outer rectrix cream-buff; inner web edged with whitish;
breast honey-yellow; feathers of throat with bases white, washed with
honey-yellow; color of under parts changing to cinnamon-buff on
abdomen, sides and under tail coverts; bill (in dried skin) brown.

The remaining specimen is marked questionably as an immature
male. It resembles the female in general, but has the dark, adult
plumage appearing on the upper surface. Beneath it is paler than the
female, and has the throat almost white. The bill is black as in the
adult males.

This fine species may be distinguished readily from Myiagm vani-
korensis (Quoy and Gaimard), to which it is allied, by its much darker
coloration, and larger, heavier bill.

SYLVIIDAE.

64. Conopoderas atypha, sp. nov.

Characters. — Similar to Conopoderas caffra (Sparrman) but upper
parts duller brown, pale margins on feathers of dorsal surface nearly
obsolete, wing much shorter, bill shorter.

Tijpe.— V. S. N. M. 212,493. Male. Polynesia: Paumotu Is-
lands; Fakarava, 11 October, 1899. C. H. Townsend.

Range. — The Paumotu Islands, Polynesia (specimens examined
from the following islands: — Whitsunday, Akiaki, Makemo, Apataki,
Fakarava, Carlshoft', Tikei, Rangiroa, Makatea, and Hereheretue).

Remarks. — This distinct species is represented by a series of forty-
two specimens. In this material there are six forms (including the
typical one) that may be considered as subspecies.

As has been pointed out by Oberholser (Proc. U. S. N. M., 1905,
28, p. 900) Turdus longirostns of Gmelin (Syst. nat., 1789, 1, pt. 2,
p. 823) based on Latham's long-billed thrush from the Island of Eimeo,
is preoccupied by Sitta caffra of Sparrman (Mus. Carlson., 1786, fasc. 1,
no. 4, pi. 4). As Sparrman assigns no locality for his bird the type-
locality of Sitta caffra Sparrman is hereby designated as Tahiti, as
this is the probable place from which his specimens came.

Conopoderas atypha apparently occurs upon most if not all of the

TOWNSEND AND WETMORE : THE BIRDS. 207

islands in the Paumotu Archipelago. With series from all the locali-
ties several forms in addition to those described here will be found
without doubt. There is much variation in this species in specimens
from the same localities. Specimens that have the entire plumage
strongly suffused with rufescent color are common, and a very gray
phase is also evident. In addition some specimens are more or less
albinistic, and there is considerable individual variation in length of
wing and tail without reference to locality. All these are confusing,
and the actual divisions and relationships among the birds from
different islands are evident only after careful study and comparison.

In the present study of this species no attempt is made to define the
form inhabiting the isolated island of Hereheretue, as the only speci-
men taken there was preserved in spirits, but there is little question
but that it is distinct. The delicate grays, browns, and yellows found
in the plumage of Conopodcras aiijpha are liable to injury from immer-
sion in alcohol, so that this bird is not available for color comparison.
Measurements of the specimen are as follows: — wing 87.0, tail 84.0,
exposed culmen 20.0, tarsus 29.5. It will be seen from these that the
tail is longer than in any other specimen available at this time, while
the culmen and tarsus are short when compared with the length of
wing. Apparently the bird was pale below and quite brown above.
To assign a subspecific name to this specimen under the circumstances
would lead to confusion, so that it is merely listed here under the
general discussion of the species.

Following are the subspecific forms of Conopodcras atypha at present
recognized.

64. CONOPODERAS ATYPHA ATYPHA.

Description. — Type p. 206. Crown, hind neck, back, and scapu-
lars olive-brown, the feathers of back edged more or less with paler;
rump between honey-yellow and Isabella color; upper tail coAcrts
buffy brown; rectrices olive-brown, the outer pair edged with whitish
and all save the two middle pairs tipped with whitish; remiges olive-
brown, the feathers margined with tilleul-buff, tertials both margined
and tipped with tilleul-buff; wing coverts olive-brown, the greater
coverts tipped obscurely with dull ivory-yellow; spot on lower eyelid
and superciliary stripe (extending forward to base of bill) iA'ory-
yellow; loral feathers fuscous, tipped with ivory -yellow; streak
behind eye olive-brown ; throat and breast whitish, washed lightly

208 bulletin: museum of comparative zoology.

with cartridge-buff; center of abdomen whitish; sides, flanks, bend
of wing, under wing coverts and under tail coverts, dull pinkish buff;
tibia pinkish buff.

Measurements. — Males (nine specimens), wing 82.5-S9.6 (86.6); tail
70-80 (74.3); exposed culmen 19-22 (20.3); tarsus 29.5-31.5 (30.6).
Female (1 specimen), wing 88.0; tail 75.5; exposed culmen 19.2;
tarsus 27.6.

Range. — Fakarava, Carlshoff, and Tikei Islands, Paumotu Islands.

Remarks. — Two males and one female were secured at Tikei,
9 October, and six males and one alcoholic specimen were collected
at Fakarava, 11 October. Specimens from Tikei are very slightly
darker above than those from Fakarava. One bird from Carlshoff
Island (secured by Peale) is more buffy below\ The present material
does not serve as sufficient grounds for recognizing these differences by
name.

65. CONOPODERAS ATYPHA RAVA, Subsp. nOV.

Characters. — Similar to ConojJodcras atyplia ati/pha but under
surface washed with massicot-yellow, under tail coverts averaging
paler, less buffy, and superciliary stripe massicot-yellow, most obscure
anteriorly.

Ty23e.~ U. S. N. M. 212,516. Male. Polynesia : Paumotu Islands ;
Whitsunday Island, 31 October, 1899. C. H. Townsend.

Description. — Crown, hind neck, back, and scapulars between
buffy brown and olive-brown; rump and upper tail coverts buft'y
brown, the rump washed with deep olive-buff; rectrices, remiges, and
wing coverts olive-brown; outer pair of rectrices margined lightly,
and three outer pairs tipped slightly, with paler; greater wing coverts
and inner tertials with paler tips and margins; primaries and second-
aries margined with deep olive-buff"; spot on lower eyelid and super-
ciliary stripe above eye massicot-yellow, the superciliary becoming
obscure as it passes forward to base of bill; loral feathers fuscous with
pale tips; auricular region dark olive-buff'; under surface washed
strongly with massicot-yellow, sides and flanks washed with brownish
cream-buff; under tail coverts between pale olive-buff and olive-buff;
bend of wing and under wing coverts washed with cream color; tibia
olive-buff.

Measurements. — Males (four specimens) wing 84.5-90.0 (87.6);
tail 72.0-76.0 (74.8); exposed culmen 18.5-20.5 (19.6); tarsus 29.2-
30.7 (29.8).

TOWNSEND AND WETMORE : THE BIRDS. 209

Female (one specimen) wing 83.5; tail 67.0; exposed culmen 20.5;
tarsus 31.0.

Range. — Whitsunday and Akiaki Islands, Paumotu Islands.

Remarks. — A male was taken on Akiaki, 30 October, and three
males and one female on Whitsunday Atoll 31 October. The dis-
tinctly yellow cast of the under parts serves to distinguish this form
from the other subspecies of Conopoderas atijpha at present known.
There is available only one bird from Akiaki and examination of more
material will show without doubt that birds from this island are
readily separable from those from Whitsunday. The single specimen
at hand, a male, differs from birds from Whitsunday in being distinctly
more buffy below, especially on the lower tail coverts. In addition
it has the rump paler and the tail more extensively tipped with white.

66. Conopoderas atypha crypta, subsp. nov.

Characters. — Similar to Conopoderas atypha atypha but distinctly
grayer above, and whiter, less buffy, below.

f?/p6!.— U. S.N. M. 212,507. Male. Polynesia: Paumotu Islands ;
Makemo, 20 October, 1899. C. H. Townsend.

Description. — Crown, hind neck, back, and scapulars hair-brown,
the feathers with paler tips and edgings; riimp dull avellaneous;
upper tail coverts drab; rectrices, remiges, and greater wing coverts
olive-brown; rectrices with slightly indicated pale margins, the four
outer pairs tipped with whitish; lesser and middle wing coverts
drab, margined indistinctly with pale drab-gray; primaries and
secondaries margined with pale drab-gTay; spot on lower eyelid and
superciliary stripe above eye ivory-yellow, the superciliary more or
less obscure anteriorly; loral feathers fuscous with pale tips; auri-
cular region between deep and dark olive-buff, becoming dark olive-
buff posteriorly; throat and breast whitish washed with marguerite-
yellow; this color becoming gradually buffy posteriorly until the
under tail coverts are light buflF; bend of wing, under wing coverts,
sides and flanks light buff.

Measurements. — Males (three specimens) wing 89.5-90.0 (89.8);
tail 77.0-80.0 (78.1); exposed culmen 20.2-21.0 (20.5); tarsus 30.0-
31.0 (30.6).

Range. — Makemo Island, Paumotu Islands.

Remarks. — Three males, one specimen with sex not indicated, and
two birds preserved in alcohol, were secured on Makemo, 20 October.

210 bulletin: museum of comparative zoology.

One male in the series examined has a strong rufescent wash on the
entire plumage, the reddish suffusion being especially marked on the
rump and posterior under parts. The others agree in being uniformly
of a grayish cast.

67. CONOPODERAS ATYPHA AGASSIZI, subsp. nov.^

Characters. — Similar to Conopodcras atypha crypta but back slightly
grayer, crown distinctly darker than the back.

Type.— U. S. N. M. 212,491. Male. Polynesia: Paumotu Islands;
Apataki, 7 October, 1899. C. H. Townsend.

Description. — Crown slightly grayer than olive-brown; hind neck,
back, scapulars, and upper tail coverts between drab and hair
brown; rump between smoke-gray and light grayish olive; rectrices
and remiges olive-brown; lesser and middle wing coverts drab;
greater wing coverts olive-brown, each feather margined broadly with
drab; rectrices and remiges margined lightly with drab or light drab;
no light tips evident on rectrices; spot on lower eyelid, and super-
ciliary stripe above and behind eye olive-buff, the superciliary more
or less obscure anteriorly; loral feathers fuscous with obscure paler
tips; auricular region between drab and hair-brown, paler anteriorly;
under surface whitish; throat, middle of breast, and upper abdomen
washed lightly with pale olive-buff; lower abdomen and under tail
coverts whitish; sides of breast becoming strongly grayish; under
wing coverts whitish; bend of wing marguerite-yellow.

Measurements. — Male (one specimen. Type) wing 89.0; tail 77.7;
exposed culmen 19.0; tarsus 28.0.

Range. — Island of Apataki, Paumotu Islands.

Remarks. — One specimen, the type, was secured on Apataki, 7
October. This bird is nearest the form inhabiting Makemo Island,
but differs from it in several particulars, none of which come within
the range of individual variation as shown in the entire series of
Conopoderas atypha examined. The crown is distinctly darker than
the back, and in addition the under wing coverts and under -tail coverts
are much paler.

68. Conopoderas atypha nesiarcha, subsp. nov.

Characters. — Similar to Conopoderas atypha atypha but bill shorter;
coloration above averaging more brownish, below more hea\'ily
suffused with buff,

1 Named in honor of Alexander Agassiz.

TOWNSEND AND WETMORE: THE BIRDS. 211

Xype.— \]. S. N. M. 212,511. Male. Polynesia: Paumotu Is-
lands; Rangiroa, 21 September, 1899. C. H. Townsend.

Description. — Crown, back, and scapulars between buffy brown
and olive-brown; hind neck buffy brown; rump dresden-brown ;
upper tail coverts Saccardo's umber; rectrices and remiges olive-
brown (somewhat faded); lesser wing coverts buft'y brown; middle
and greater wing coverts olive-brown, margined with buffy brown,
remiges edged with buffy brown; rectrices tipped obscurely with
tilleul-buff; lores, spot on lower eyelid and superciliary stripe cream-
buff; malar region ivory-yellow; auricular region drab, with a wash
anteriorly of olive-yellow; under surface pale olive-buff'; sides of
breast washed with buffy brown ; sides and flanks washed with Isabella
color; under tail coverts pinkish buff; bend of wing and under wing
coverts pinkish buff.

Measurements. — Males (four specimens) wing 84.2-89.5 (87.8) ;
tail 72.0-78.5 (75.0); exposed culmen 18.3-18.5 (18.4); tarsus 29.0-
31.2 (30.1).

Female (one specimen) wing 83.0, tail 70.5, exposed culmen 19.0,
tarsus 29.5.

Range. — Island of Rangiroa, Paumotu Islands.

Remarks. — Four males, one female, and a specimen preserved in
alcohol were collected on Rangiroa on 21, 22, and 23 September.
This form is more buffy below than any of the other subspecies of
Conopoderas atypha examined, save the one from Makatea (a descrip-
tion of which follows), and the short bill serves to distinguish it at
once from all others. The single female examined is grayer above and
less buffy below than the males. One male shows a tendency toward
albinism, as it has a light bar across the ends of the inner greater
coverts.

69. Conopoderas atypha erema, subsp. nov.

Characters. — Similar to Conopoderas atypha atypha but larger; bill
longer and hea\aer; coloration above brighter brown, rump more
rufous, more buffy below especially on breast, sides, and flanks ; tail more
distinctly tipped with white.

Type.— JJ. S. N. M. 212,505. Male. Polynesia: Paumotu Is-
lands; Makatea, 6 October, 1899. C. H. Townsend.

Description. — Crown and hind neck slightly darker than buffy
brown ; back and scapulars olive-brown, the feathers margined rather

212 bulletin: museum of comparative zoology.

obscurely with dull pinkish buff; rump between clay color and tawny
olive; upper tail coverts Saccardo's umber; rectrices and remiges
olive-brown; lesser wing coverts tawny olive; middle and greater
coverts olive-brown the feathers margined with pinkish buff; outer
web of outer tail feather and tips of others obscurely white; rest of
rectrices and remiges margined with pale olive-buff; superciliary
stripe and spot on lower eyelid cream-buff; superciliary distinct;
lores dull white; auricular region changing from dull cream-buff"
anteriorly to buffy brown posteriorly; throat and upper breast and
abdomen washed with ivory-yellow; lower breast washed with cha-
mois, this color deepening somewhat on the sides ; flanks washed with
cinnamon-buff; under tail coverts pinkish buff'; bend of wing pinkish
buff; under wing co-verts between pinkish buff and cinnamon-buff.

Measurements. — Males (four specimens) wing 88.2-91.5 (90.1); tail
78.2-82.0 (79.4); exposed culmen 21.0-22.5 (22.0); tarsus 30.0-33.5
(31.5).

Range. — Island of Makatea, Paumotu Islands.

Remarks. — Four specimens were secured on Makatea Island, 26
September, and two more were added to the collection from the
same locality, 6 October. Four of these birds are males; the other
two do not have the sex indicated certainly. This is the most dis-
tinct of all the forms of Conopoderas atypha at present known, a
circumstance to be expected as Makatea is cut oft' by deep ocean from
the main Paumotu Group. The central islands lie on a plane bounded
by the thousand fathom curve, while Makatea is outside of this
irregular line.

Two specimens show an indication of albinism. One has the rec-
trices, save for the two central pairs, entirely white. In the other one
the outer rectrix is white, and there are irregular white blotches on the
tips of the others. In size, and in more prominent pale margins on the
feathers of the dorsal surface this form shows a very slight approach
toward Conopoderas caffra of Tahiti. It is so distinct, however, in its
much browner coloration, smaller size, and general appearance that it
cannot be considered as a connecting link between that species and
C. atypha. C. a. erema is distinguished from all other forms of C.
atypha known at present by its larger size, longer bill, and much more
bufty plumage. The differences are in fact almost sufficient to give
it recognition as a separate species. In view of the great variation
in the wide ranging C. atypha, however, it seems better to consider
the Makatea bird as a subspecies.

TOWNSEND AND WETMORE: THE BIRDS. 213

70. CONOPODERAS PERCERNIS, Sp. nOV.

Characters. — Similar to Conopoderas rnendanae (Tristram) , but
outer web of the external rectrices dusky (with only a faint yellow
margin), more yellow below, with basal portion of the inner webs of
the primaries white very faintly tinged with yellow.

Tijpe.— V. S. N. M. 212,479. Male. Polynesia: Marquesas;
Nukuhiva, 15 September, 1899.

Description. — Feathers of crown and line behind eye. with centers
deep olive to dark olive, this color merging into a broad margin of
light yellowish olive; neck, back, and scapulars between yellowish
olive and dark greenish olive, this color changing to light yellowish
olive on the tips of the feathers on back and scapulars, on rump
changing to deep colonial-buff; upper tail coverts light brownish
olive with margins more yellowish ; feathers of wings and tail fuscous
black; primaries edged with light yellowish olive; secondaries edged
broadly with colonial-buff and primaries and secondaries tipped with
marguerite-yellow; wing coverts fuscous, margined broadly with
colonial-buft'; rectrices tipped with marguerite-yellow, the tips
broader on the external feathers, lessening in extent, especially on the
inner webs, toward the central pair; outer pair with a very narrow
obscurely indicated paler margin for distal half; under surface, save
abdomen, barium-yellow to citron-yellow; center of abdomen white;
lores whitish; superciliary stripe, malar and auricular region, and
bend of wing strontian-yellow; under wing covers naphthalene-
yellow; inner webs of primaries for basal half white.

Measurements. — Seven males wing 98.6-103.0 (100.5) ; tail 86.2-
92.3 (88.7); exposed culmen 22.5-24.0 (23.3); tarsus 31.0-33.3 (32.4).

Range. — Island of Nukuhiva, Marquesas Islands.

Remarks. — This species is represented by nine skins, two mummies,
and an alcoholic specimen secured at Nukuhiva, 15, 16 September,
1899. The series is remarkably constant in coloration, there being
no tendency toward albinism such as is often found in the island
inhabiting species of this genus. The seven birds that have the sex
determined are males.

No specimens of Conopoderas niendanae (Tristram) are aA'ailable for
comparison, but an excellent description is found in the Catalogue
of birds in the British museum, 1883, 7, p. 526, and with the original
description of Canon Tristram (Ibis, 1883, p. 43) is given a colored
figure that shows the characters of the bird distinctly. In this plate

214 bulletin: museum of comparative zoology.

of C. mendanae the yellow of the outer tail feathers is especially notice-
able. Twelve specimens in all of Conopodcras percernis have been
examined, and all are constant in the characters assigned in the
diagnosis above.

71. CONOPODERAS SYRINX (Kittlitz).

Sylvia syrinx Kittlitz, Mem. Acad, imper. sci. St. Petersburg, 1835, 2, livr. 1,
p. 6, pi. 8. (Lugunor and Uleei).

One female was taken at Ponape, Eastern Carolines, and one male,
two females and an immature bird (mummy) come from Uala in the
Truk Group, collected 16 February, 1900. There are two forms
indicated in this material, but as specimens that may be considered
typical C. syrinx of Kittlitz are not available for comparison it is not
advisable to separate them at this time. Kittlitz described this bird
from the Lugunor and Uleei Groups. Finsch (Journ. Mus. Godeffroy,
1876, 12, p. 30) says that Kittlitz also included birds from Ualan.
Specimens from none of these localities are available in the U. S. N. M.
collections. The birds from Uala, together with a specimen marked
Ruk collected by Kubary, differ from the bird from Ponape in being
distinctly paler. The head and neck are less brownish, especially
on the sides of the neck, and are much grayer than the back. The
rump and upper tail coverts are paler, and the under parts are less
extensively cinnamon-buff especially on the sides and under tail
coverts. Birds from the two islands agree in measurements.

CONOPOPHAGIDAE.

72. Lalage pacifica (Gmelin).

Turdus pacificus Gmelin, Syst. nat., 1789, 1, pt. 2, p. 813. (Friendly Islands).

Three males were taken at Eua, 28 November, two males at Tonga-
tabu, 30 November, a female at Nomuka, 2 December, and two males
at Vavau, 4 December, all in the Tonga Group. In addition a female
was secured at Kambara, 7 December, and a male in immature
plumage at Viti Levu 18 December, in the Fiji Islands.

Birds from Eua have the rump slightly paler and average a trifle

TOWNSEND AND WETMORE: THE BIRDS. 215

larger than others but the differences are too sHght to be dependable
in such a small series. Average measurements of males from the
Tonga Islands follow.

Sex

Locality

Wing

Tail

Exposed
Culmen

Tarsi,

2 d'd'

Vavau

98.7

66.7

13.5

25.9

2 cfc^

Tongatabu

99.0

66.5

13.5

25.9

3 d^cf

Eua

99.2

67.1

14.0

27.1

73. Lalage whitmeei Sharpe.

Lalage whitmeei Sharpe, Mittheil. K. zool. mus. Dresden, 1878, p. 371. (Savage
Island).

Four males and two females were secured at Niue, or Savage Island,
25 November. Both of the females are immature. Each shows
a strong wash of bright brown on the lesser and middle wing coverts.

LANIIDAE.

74. Pachycephala jacquinoti Bonaparte.

Pachycephala jacquinoti Bonaparte, Conspec. gen. Avium, 1850, 1, p. 329.
(Vavau).

Four adult males and one male in immature plumage were collected
at Vavau, Tonga Islands, 4, 5 December. The male in juvenal
plumage is duller in color than the adults, and has the crown deep
mouse-gray with a few feathers of the adult plumage showing. The
chin and throat are whitish, the --feathers tipped with deep mouse-
gray, especially laterally, and the upper breast is light drab. A few
black feathers show here also. The side of the head is light grayish
olive, and the auricular region wood-brown. The light ring on the
hind neck is barely indicated, and the under parts are much paler
yellow.

Gray (Gen. birds, 1845, 1, p. 271) lists this form under Pachycephala
and refers to the plate published by Hombron and Jacquinot (Voy.
Pole Sud. Zool. Oiseaux, October, 1844, pi. 5, fig. 2) where the bird
is called "Pie-grieche a masque noir" but does not designate a name
for it.

216 bulletin: museum of comparative zoology.

PRIONOPIDAE.

75. Pinarolestes heinei (Finsch and Hartlaub).

Myiolestis heinei Finsch and Hartlaub, Proc. Zool. soc. London, 1869, p. 546.
(Tonga Islands) .

Two males and three females were secured at Nomuka, 2 Decem-
ber. The Ught margin of the edges of the mandibles seems very
characteristic in this species. The sexes are similar in color. There
seem to be no pre\'ious records of the species from this island.

76. Pinarolestes nesiotes, sp. nov.

Characters. — Similar to Pinarolestes mtiensis (Hartlaub) but
larger, with heavier bill, and with white tips absent on four central
rectrices and more restricted on the remaining feathers.

Type.— J]. S. N. M. 212,438. Female. Polynesia: Fiji Islands;
Kambara, 7 December, 1899. C. H. Townsend.

Description. — Crown and hind neck mummy-brown, changing to
raw umber on the back and rump; wing and tail fuscous, the wing
coverts, outer webs of primaries and edgings of secondaries raw
umber; rectrices save the two central pairs lightly tipped with white;
four central tail feathers immaculate, with darker cross bars faintly
indicated in certain lights; throat, breast, and abdomen smoke-gray;
feathers of forehead with smoke-gray bases, forming an obscure
patch of pale color; eye ring and lores obscurely blackish; auricular
region hair-brown; rami, side's of head behind eye, sides of neck, and
sides of breast light grayish olive; sides of abdomen and flanks washed
with cinnamon-brown; under tail coverts buffy brown; tibia mouse-
gray.

Measurements. — -Female {Type) wing 95; tail 80; exposed culmen
21; tarsus 25.

Range. — Island of Kambara, Fiji Islands.

Remarks. — One female (the Type) was secured at Kambara, Fiji
Islands, 7 December, 1899. The genus does not seem to have been
known from this island previously. Study of a large series of Pinaro-
lestes of the mtiensis group will probably show that the bird described
here is a subspecies of vitiensis. As given by Sharpe (Cat. birds Brit,
mus., 1877, 3, p. 300) the wing of P. vitiensis (sexes not indicated)

TOWNSEND AND WETMORE: THE BIRDS. 217

varies from about 79 to 86 millimeters. As females in this genus are
smaller than males it appears that the bird described here is a very
large form. Pinarolcstes nesiotes appears to resemble P. comyres-
sirostris (Layard), as it has a large and somewhat compressed bill,
but differs from that form (judging from descriptions) in larger size,
and shorter bill, while in addition it lacks white tips on the central
tail feathers.

STURNIDAE.

77. AcRiDOTHEREs TRiSTis (Liune).

Paradisea tristis Linn6, Syst. nat., ed. 12, 1766, 1, p. 167. ("Philippines.")-

A male of this species was collected on Viti Levu, in the Fiji Islands,
14 December. There seem to be no previous records of this species
here, and no information is available at present to show when it was
introduced.

EULABETIDAE.

78. Aplonis tabuensis (Gmelin).

Lanius tabuensis Gmelin, Syst. nat., 1788, 1, pt. 1, p. 306. (Friendly Islands) .

Of seven specimens that were collected four were taken on Eua, 28
November, two on Nomuka, 2 December, and one on Vavau, 4 Decem-
ber, all in the Tonga Group. One does not have the sex indicated,
the others are all males. Specimens from the three localities agree
closely. The single bird from Vavau has the tarsus longer, and is
somewhat clearer gray below than the others, while birds from Eua
have the dark mark on either side^at the base of the mandible slightly
more prominent. Otherwise the specimens in the series are very
similar. Measurements follow : —

No.

Sex

Locality

Wing

Tail

Exposed
Culmen

Tarsui

U. S. N. M. 212,.373

Eua

113.0

63.0

20.5

27.0

212,374

112.0

64.5

22.0

30.0

" 212,376

114.. 5

70.0

20.0

27.0

212,377

Nomuka

111.0

67.5

21.0

29.0

212,378

114.0

63.5

20.5

29.5

212,372

Vavau

113.5

68.5

20.0

33.0

218 bulletin: museum of comparative zoology.

79. Aplonis vitiensis Layard.

Aplonis vitiensis Layard, Proc. Zool. soc. London, 1876, p. 502. (Fiji Islands).

A female was taken on Kambara in the Fiji Islands, 7 December.
This bird is an adult in very worn plumage.

SO. Aplonis brunnescens Sharpe.

Aplonis brunnescens Sharpe, Cat. birds Brit, mus., 1890, 13, p. 132, pi. 6.
(Savage Island).

There are in the collection six skins and one alcoholic specimen
collected on Nine or Savage Island, 25 November. The sexes are
apparently similar in color, but females have the wing averaging shorter
than males. One male has a strong rufescent wash on the feathers of
the entire plumage. This wash is most pronounced on head, neck,
breast, and back. Following are measurements of the skins exam-
ined:—

Sex

Wing

Tail

Exposed
Culmen

Tarsus

U. S. N.

. 212,379

103.5

61.0

18.0

31. C

212,.380

98.0

54.0

17.0

29.0

212,381

cf ?

102.0

55.0

17.0

30.0

212,382

98.0

58.0

17.0

31.0

212,383

? ?

98.0

55.5

17.0

28.5

" 212,384 9 97.0 54.5 17.0 29.5

Oberholser (Bull. 98 U. S. N. M., 1917, p. 58, 59) has recently
pointed out the distinctions between the genus Lamprocorax Bona-
parte and Aplonis Gould, showing that in the former group the second
primary (counting from the outside, and beginning with the spurious
primary) is longer than the fifth, while in Aplonis the second primary
is sometimes equal to but usually shorter than the fifth. Aplonis
hrunncsccns was not included in his list of species studied in this
connection, but upon examination is found to have the second primary
shorter than the fifth so that this species is a true Aplonis. Aplonis
cinerascens Hartlaub and Finsch also belongs here, as it has the more
rounded wing found in this group.

TOWNSEND AND WETMORE : THE BIRDS. 219

81. Aplonis opaca (Kittlitz).

Lamproth[ornis] opaca Kittlitz, Kupfertafeln naturgesch. Vogel, 1833, 2,
p. 11, pi. 15, fig. 2. (Marianne and Caroline Islands).

Nine specimens of this bird were secured in the Caroline Islands.
Four males and one female were taken on Kusaie, 8 February, one
male on Ponape, 11 February, and a female on Uala in the Truk
Group, 16 February. In addition an alcoholic specimen (preserved
now as a mummy) was taken on Kusaie 9 February, and there are
two skins marked from the Caroline Islands with no definite locality
given.

These birds all seem to be immature, and none show iridescence on
the feathers to any great degree. These specimens average smaller
than a series from Guam in the collection of the U. S. N. M., but as
immature birds in the same stage of plumage are lacking from the
Ladrone Islands specimens from the two localities are not directly
comparable. Oberholser (Bull. 98 U. S. N. M., 1917, p. 59) has
shown that the name Lamprothornis opaca given by Kittlitz as a
manuscript name of Lichtenstein is available for this present species.

Measurements of the specimens from the Caroline Islands follow: —

Sex

Locality

Wing

Tail

Exposed
Culmen

Tarsui

U. S. N,

.M.

212,363

Ponape

121.0

84.0

23.5

31.0

212,364

Kusiae

111.0

21.5

28.5

212,365

116.0

74.5

21.0

30.5

212,366

121.5

79.0

22.5

30.5

212,368

123.5

81.5

23.0

30.0

212,362

Uala

116.5

66.0

20.5

31.5

212,367

Kusaie

121.0

75.0

21.0

30.0

MELIPHAGIDAE.
82. Myzomela rubratra rubratra (Lesson).

Cinnyris rubraier Lesson, Diet. sci. nat., 1827, 50, p. 30. ("Oualan").

Four males, three females, and an immature bird secured on Kusaie
in the Middle Carolines, 8, 9 February were made into skins and in
addition eight birds were preserved as mummies. Males and females

220 bulletin: museum of comparative zoology.

are similar in color, but the latter have the wings and tail somewhat
shorter. Adults seem to vary in depth of color with age. The imma-
ture bird is dull black above and brownish below while red feathers
are beginning to show on the back and about the head.

Two male birds secured on Uala, 16 February, are not quite
typical of this form. The red of the plumage is slightly paler, ap-
proaching M. r. dichromata from Ponape. As these specimens from
Uala agree otherwise with the large series from Kusaie they are placed
with that form for the present. With a larger series more striking
differences may be apparent.

Cinnyris rubrater of Lesson is usually quoted as from Voy. Coq.

1826, p. 678. According to Sherborn and Woodward (Ann. mag. nat.
hist, 1901, ser. 7, 7, p. 391) Voy. Coq., 1, pt. 2, p. 678, should date
from 1830. This makes the first reference date from the Diction-
naire des science naturelles, 50, p. 30, as this volume appeared in

1827. The description there given is practically identical with that
appearing under the later reference. In his description Lesson says,
"cette espece, qui existe au Museum d'histoire naturelle, habite les
lies Philipines, oil I'a trouvee M. Dussumier, et I'ile d'Oualan, oil j'en
ai tue un grand nombre d'individus." In assigning it to the Philip-
pines Lesson was in error as the species is confined to the Ladrone,
Pelew, and Caroline Islands. His statement may have basis in the
fact that vessels proceeding to the Philippines in the old days usually
touched at Guam, so that M. Dussumier may have seen and collected
the species there. From this Lesson may have attributed statements
or specimens emanating from M. Dussumier to the Philippine Islands,
as it was there that that naturalist carried on the major part of his
work. The type-locality must stand as Ualan (known now as Kusaie)
from which the large series in the present collection came.

83. Myzomela rubrata dichromata, subsp. nov.

Characters. — Similar to Myzomela rubrata rubrata (Lesson) but
black markings of head deeper in color and more extensive, co^'ering
orbital ring, lower eyelids, antorbital and loral region, and extending
around the base of the bill on forehead and chin, red of plumage
slightly paler.

TyiJc.— V. S. N. M. 212,413. Adult male. Polynesia: Eastern
Caroline Islands; Ponape, 11 February, 1900. C. H. Townsend.

Description. — Wings, scapulars, a broken band across shoulders.

TOWNSEND AND WETMORE : THE BIRDS. 221

tail, thighs, under wing coverts, lores, antorbital region, orbital ring,
lower eyelids and anterior portion of forehead and chin black, the
latter connecting with the loral region so that there is a continuous
ring of black around the bill; under tail coverts and lower abdomen
slightly duller black; rest of plumage including most of head, back,
tips to some of scapulars, upper tail coverts, throat, breast, and upper
abdomen between scarlet-red and scarlet.

Measurements. — Males (two specimens) wing 76.0-79.0; tail
50.0-52.5; exposed culmen 18.0-18.5; tarsus 22.0.

Female (one specimen) wing 70.0; tail 45.0, exposed culmen 16.0;
tarsus 20.5.

Range. — Island of Ponape, Eastern Caroline Islands, Polynesia.

Remarks. — An adult and an immature male, an immature female,
and an adult specimen preserved as a mummy were collected on
Ponape in the Eastern Carolines, 11, 12 February. This form,
though paler than Myzomela r. rvbrata (Lesson) from Kusiae, the
type-locality, has the red in its plumage darker than the color found
in Myzomela r. saffordi Wetmore (Proc. Biol. soc. Washington, 1917,
30, p. 117) from Guam. The immature plumages show nothing
remarkable.

The recognized forms of the red and black honey-eaters with
their ranges as known at present are as follows: —

Myzomela rubrata rubrata (Lesson). Kusaie, Uala (not typical).
This form probably occurs elsewhere in the Middle and Eastern
Carolines.

Myzomela rubrata dichromata, Wetmore. Ponape.

Myzomela rubrata saffordi Wetmore. Guam, Saipan (speci-
mens from other islands in the Ladrone Group not seen).

84. Myzomela nigriventris Peale.

Myzomela nigriventris Peale, U. S. explor. exped., 1848, 8, p. 150. (Samoan
Islands).

One bird, an adult, from Upolu, Samoan Islands, with date of col-
lection not marked was preserved as a mummy. Gadow (Cat. birds
Brit, mus., 1884, 9, p. 130) indicates that Myzomela nigriventris is a
subspecies of M. cardinalis. The material at hand is not adequate
for proper comparison, in default of which M. nigriventris is for the
present accorded the rank of a species.

222 bulletin: museum of comparative zoology.

85. Myzomela jugularis Peale.

Myzomela jugularis Peale, U. S. explor. exped., 1848, 8, p. 150. (Feejee Islands).

Three males and one other specimen with sex not marked were
secured on Kambara, 7 December, and two mummies taken 16 Decem-
ber, were preserved from Viti Levu. Both locaUties are in the Fiji
Group. Birds from Kambara are sUghtly larger than specimens
from other localities, but the series available is too small to show
reliable differences between insular forms.

Measurements of specimens with sex indicated are as follows: —

No. Sex Locality Wing Tail _ , Tarsus

Culmen

M. C.Z. 81,987 (U.S.N.M. 212,420) d' Kambara 58.0 38.0 16.5 17.0

212,421 (f " 60.0 37.0 16.5 17.5

M. C. Z. 81,988 ( " 212,422) d" " 60.0 39.0 17.0 17.5

86. Meliphaga carunculata (Gmelin).
Cerihia carunculata Gmelin, Syst. nat., 1788, 1, pt. 1, p. 472. (Tongatabu).

Eleven specimens of this species were collected. The localities
and dates of collection are as follows : — two males, one female, and
one bird with sex not indicated from Eua, 28 November; two males
from Tongatabu, 30 November; three males and one female from
Nomuka, 2 December, all from the Tonga Group, and an immature
bird from Upolu, Samoan Islands, with sex and date not marked.
The specimens from the Tonga Islands seem very uniform. Birds
from Nomuka are slightly larger than the others, though in very
worn plumage. Two specimens from Eua are immature birds in
fresh bright plumage. The specimen from Upolu is not yet fully
grown. In it the cheek wattles are small, and the head is paler than
in others.

Measurements of these birds follow : —

Sex

Locality

Wing

Tail

Exposed
Culmen

Tarsi

U. S. N,

.M,

, 212,388

Eua

100.0

85.0

23.0

32.0

212,389

cf im.

100.5

84.5

23.5

31.0

212,394

Tongatabu

103.5

88.0

25.0

32.5

212,-395

100.5

84.0

31.5

212,390

Nomuka

105.5

88.0

23.0

.30.0

212,.392

105.0

85.0

23.5

30.0

^12,393

108.0

86.5

23.5

31.0

212,386

Eua

94.0

78.5

21.0

29.5

212,391

Nomuka

92.5

75.5

21.0

28.0

TOWNSEND AND WETMORE : THE BIRDS. 223

87. Meliphaga procerior (Finsch and Hartlaub).

Ptilotis procerior Finsch and Hartlaub, Beitr. fauna Central-Polynesiens, 1867,
p. 62, pi. 5, fig. 1. (Ovalau).

Four specimens were secured on Viti Levu in the Fiji Islands. One
is a male, another is marked male with a query, and on the labels of
the other two birds the sex is not indicated. An immature bird not
yet fully grown has the bare malar space much reduced in size, and
the black markings about the head are duller, but otherwise the bird
is similar to adults. Wiglesworth (Aves Polynesiae, 1891, p. 34-35)
has divided this species into several subspecies of which these birds
should represent the typical form. As other specimens are not
available for comparison no attempt is made here to distinguish
between forms from the various islands.

Measurements of three adults follow: —

No.

Sex

Wing

Tail

Exposed
Culmen

Tarsu

. S. N. M. 212,397

94.5

79.5

25.5

29.0

212,398

c? ?

95.0

79.5

24.0

29.0

212,399

100.0

82.5

24.5

30.0

ZOSTEROPIDAE.

88. Zosterops semperi owstoni Hartert.

Zosterops semperi owstoni Hartert, Nov. zooL, 1900, 7, p. 2. (Ruk).

Four females of this species were secured at Uala in the Truk
Group, Middle Carolines, 16 February. These vary somewhat in the
depth of black on the lores and under the eye, but this may be due
to age. The ear coverts are only slightly paler than the crown.
Measurements of these specimens are as follows: —

No.

Sex

Wing

Tail

Exposed
Culmen

Tarsus

U. S. N. M.

212,429

53.0

35.6

9.5

19.0

212,430

52.3

35.6

10.4

20.0

212,431

54.3

36.0

9.2

19.0

M. C. Z. 81,986 (

212,432)

54.0

35.2

10.0

20.4

224 bulletin: museum of comparative zoology.

89. ZosTEROPS FLAViCEPS Peale.

Zosterops flaviceps Peale, U. S. explor. exped., 1848, 8, p. 95. (Venua-levu) .

A single specimen collected on Viti Levu, 16 December, is preserved
as a mummy. This bird has been called Z. flaviceps provisionally, as
specimens of allied races from Australia and elsewhere are not at
hand so that it is not possible to settle the true status of this form.
It is probable that eventually it may be established as a subspecies
of Z. lateralis (Latham). The present specimen resembles closely
the type of Z. flaviceps, which is preserved in the U. S. N. M., and
agrees with the description of that form in having the under tail
coverts distinctly yellow. It has the following measurements : — wing
56.0; tail 45.0; exposed culmen 12.0; tarsus 18.3.

90. ZoSTEROPS CINEREA (Kittlitz).

Drepanis cinerea Kittlitz, Kupfertafeln naturgesch. Vogel., 1832, 1, p. 6,
pi. 8, fig. 2. (Ualan).

Two males were collected on Kusaie, Eastern Carolines, 9 Feb-
ruary. These measure as follows: —

No. Wing Tail , Tarsus

culmen

M. C. Z. 81,985 (U. S. N. M. 212,426) 63.0 39.0 13.0 20.5

" 212,427) 62.0 36.2 12.5 20.0

One of these specimens still has the tongue in place. The tip is
divided into a number of filaments, all fine, but varying in size, that
tend to turn somewhat in slow spirals.

91. ZosTEROPS PONAPENENSis Finsch.

Zosterops ponapenensis Finsch, Journ. mus. Godeffroy, 1876, 12, p. 27, pi. 2,
fig. 1. (Ponape).

One male was collected on Ponape in the Eastern Carolines,
1 1 February. Besides being more brownish in coloration this species
is smaller than Z. cinerea. The bill in particular is shorter. Meas-

TOWNSEND AND WETMORE: THE BIRDS. 225

urements of this specimen are as follows: — wing 59.2; tail 38.2;
exposed culmen 11.0; tarsus 20.0.

Though the plate given by Finsch is marked po7iapensis, in the
original description of this bird the name is given as Zosterops pona-
penensis.

PLOCEIDAE.

LONCHURA CASTANEOTHORAX (GoULD).

Amadina castaneothorax Gould, Synop. birds Australia, 1837, pt. 2, pi. 21.
(Cairns, Queensland).

A bird marked questionably as a female was taken at Tahiti in the
Society Islands, 3 October. It has the under tail coverts white
with brownish bases. An adult male was secured at Bora Bora in
the same group, 17 November. This introduced species does not
seem to have been recorded previously.

93. Aegintha temporalis (Latham).

Fringilla temporalis Latham, Ind. ornith. Suppl., 1801, p. xlviii. (New South
Wales).

A male was secured at Tahiti, in the Society Islands, 13 Novem-
ber, and two females were collected the following day. This is an
introduced form that does not seem to have been recorded here pre-
viously.

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE

Vol. LXIII. No. 5.

THE PSAMMOCHARIDAE OF
WESTERN NORTH AMERICA.

By Nathan Banks.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

September, 1919.

No. 5. — The Psammocharidae of ]]'cstcrn North America.

By Nathan Banks.

The Psammocharidae, a family of fossorial H\Tnenoptera, long
known as Pompilidae, are mostly black insects of moderate size;
the females are provided with a \'ery painful sting, doubtless as severe
as that of any hymenopteron. A few species are large with bright
reddish or yellowish wings, others are small, pale colored, but as a rule
there is little variation in color. They are very active and not easily
taken. Owing to their agility, stinging ability, and dull colors, they
have been more neglected by collectors and systematists than other
families of fossorial Hymenoptera.

In the past few years I have received collections from Messrs. R. W.
Doane, E. P. and M. C. Van Duzee, and especially C. F. Baker;
recently a large amount of Western material from Prof. J. Chester
Bradley collected mostly by himself. In the M. C. Z. is much mate-
rial collected by Mr. Henshaw in Washington in 1882.

Compared with the Eastern States the fauna presents more of the
Sophropompilus and Aporinellus, and less of Pseudagenia and Priocne-
mis, but the most notable fact is the absence of red-banded black
species. These are very common in the East while I have seen but
one from the West (Washington).

Sy7iopsis of Genera.

1. Claws of hind tarsi bent at right angles; antennae situate considerably

above the clypeus; lab rum exserted for its entire length; metanotum

at base bilobed ^ Ceratopales.

Claws not bent at right angles 2

2. A transverse furrow on the second ventral segment (absent in some males) ;
in the fore wings the second discoidal cell at base is right-angled, and

without a distinct pocket 3

No such furrow on the venter ; in fore wings the second discoidal cell has a
small pocket at base, the hind tibiae never serrate, always more or less
irregularly spined 8

3. Hind tibiae without spines or only very weak ones, never serrate; in hind

wings the transverse median vein ends before the cubitus; last tarsal

joint without spines beneath 4

Hind tibiae more or less serrately spined, if (male) nearly smooth, then the
transverse median vein of hind wings is not before the cubitus 5

230 bulletin: museum of comparative zoology.

4. Metanotum with erect hair above Pseudagenia.

Metanotum without erect hair above Ageniella.

5. Last joint of hind tarsus without spines beneath Priocnemis.

Last joint of hind tarsus with distinct spines beneath 6

6. In the fore wings the first recurrent vein meets the second submarginal

cell before or at the basal third Pepsis.

In the fore wings the first recurrent vein meets the second submarginal
cell beyond the basal third 7

7. First recurrent vein meets the second submarginal cell close to the tip

of that cell Mygnimia.

First recurrent vein meets the second submarginal cell near the middle.

Cryptocheilus.

8. Pronotum longer than the mesonotum, nearly flat above, scarcely arched

longitudinally, last joint of hind tarsus without spines beneath, tarsus

I of female without comb of spines 9

Pronotum shorter than the mesonotum, plainly arched longitudinally. 11

9. With two submarginal cells Planiceps.

With three submarginal cells 10

10. Transverse vein in fore wings sloping backwards; ocelli placed unusually

low Psorthaspis.

Transverse vein in fore wings normal, as also the ocelli. . . .Pedinaspis.

11. Basal abdominal segment with appressed pubescence different from that

on the other segments Episyron.

Without such pubescence 12

12. Metanotum produced angularly at the posterior corners; usually but

two submarginal cells Aporinellus.

Metanotum not so produced; three submarginal cells 13

13. Metanotum transversely striate Ridestus.

Metanotimi not striate 14

14. A short impressed line or groove on posterior part of the pronotum . . 15
No such line or groove 16

15. Metanotum distinctly grooved at base; upper margin of clypeus nearly

evenly curved Arachnophroctonus.

Metanotum not grooved at base; clypeal suture sinuate or zigzag.

Batazonus.

16. Third joint of antennae of female barely if any longer than the first joint,

spines under last joint of hind tarsus very weak 17

Third joint of antennae plainly longer than first 18

17. Metanotum hairy Sophropompilus.

Metanotum not hairy Nannopompilus.

18. Third abdominal segment hairy above as well as rest of the body and

femora Arachnophila (Alaska).

Third segment not hairy above 19

19. No spines under last joint of the hind nor front tarsi, clypeus not emargi-

nate, third submarginal cell not petiolate 20

Distinct spines under last joint of the hind tarsi; no malar space 22

banks: psammocharidae. 231

20. A distinct malar space between eyes and base of mandibles 21

No such space; metanotum not hairy Gymnochares.

21. A comb to front tarsi of female; metanotum oblique; small species.

Agenoideus.
No comb; metanotum flat at base; larger black species. . Alloc ij phony x.

22. Clj^jeus of female strongly emarginate in middle; pronotum arcuate

behind ; a comb to front tarsi Lophopompilus

CljTDeus not emarginate 23

23. A strong comb to front tarsus of female; metanotum hairj' above.

Psatnmochares.
No comb to front tarsus 24

24. Metanotum not hairy above; marginal cell much more than its length

from tip of the wing Pompiloides.

Metanotum plainly hairy above; marginal cell more acute at tip, and
nearer to wing tip Anoplius.

Lophopompilus.

Four long spines in comb on first tarsal joint cleora.

Three rather short spines, hardly forming a comb on first tarsal joint, .aethio'ps.

Lophopompilus aethiops (Cresson).
Calif.: Exeter, 30 July; Washington (Kincaid).

Lophopompilus cleora Banks.

Calif. : Dyerville, July, Los Angeles, 3 May; Wash. : Camp Umatilla,
26 June, 1882.

Psammochares.

1. Pronotum arcuate behind; three spines in tarsal comb on first joint;

larger species bellicosus.

Pronotum angulate behind 2

2. With three spines on first tarsal joint in the comb eureka.

With four spines on first tarsal joint 3

3. Spines of comb on second joint longer than that joint scelestus.

Spines of comb on second joint shorter than that joint anoplinus.

232 bulletin: museum of comparative zoology.

1. Pronotum arcuate behind, large species hellicosus.

Pronotum angulate behind ~ 2

2. Ventral segments with tufts on last two eureka.

No tufts of hair on ventral segments scelestus.

PSAMMOCHARES ANOPLINUS, Sp. nOV.

Type.— Cornell Univ. Coll. Paratype M. C. Z. 10,396.
Alberta: Medicine Lake to Jasper, 4 July (Bradley); Calif.: Lake
Tahoe, 6,200 ft., 21 August, (Bradley); Oregon: Umatilla, 24 June,

1882.

9 Black, similar to P. scelestus in nearly all respects; in hind tarsi the
fourth joint is a little longer than in P. scelestus, being nearly equal to the fifth;
in the fore wings the basal vein is plainly a little before the transverse. The
anterior tarsus has the comb of very much shorter spines than in P. scelestus,
the two on the second joint hardly half as long as those in P. scelestus; there
are four short ones on the first joint, all these spines are rather stout and a
trifle curved.

Length 11 mm.

PSAMMOCHARES EUREKA, sp. nOV.

Type— M. C. Z. 10,397. Paratype— Cornell Univ. Coll.

Calif.: Lincoln Park, San Francisco (Giffard), Ingleside, 26 August
(Bradley), Felton, Santa Cruz Mts., 15-20 May (Bradley), Lakeside,
8 May (Bradley).

9 Black, densely long haired; clypeus broad, truncate in front, slightly
margined; first joint of antennae with hair beneath, third joint quite long,
second and third together equal to vertex width, groove above antehnae to
anterior ocellus, hind ocelli about as close to eyes as to each other, vertex from
in front straight across; pronotum angulate behind; metanotum with a median
groove, not very deep; legs moderately long, not very spiny, hind tibiae with
only about four above, inner spur of hind tibia reaches little beyond middle of
the basitarsus, three spines in comb of the first tarsal joint, the last not equal
to next joint. Wings violaceous black, marginal cell hardly its length from
wing tip, second and third submarginal cell subequal, both broad above,
second recurrent reaching cell before the middle, basal vein a little before the
transverse. Male more slender, and much smaller; the last two ventral seg-

banks: psammocharidae. 233

ments before the tip have dense brushes of erect hair; the ventral plate is
densely clothed with short hairs, and without the prominent carina of P.
scelestus.

Length 9 , 9 mm.

PSAMMOCHARES SCELESTUS (CreSSOn).

Washington (Kincaid);. Calif.: Sequoia National Park, 21-25
July, Los Gatos Divide to Mt. Diabolo, 6-8 June, Lemoncove,
Tulare Co., 9 July, Mesa Grande, Russian River, 30 September,
Harris, Humboldt Co., 29 June, Three Rivers, Tulare Co., 16 July,
Ingleside, 25 August.

PsAMMOCHARES BELLicosus Banks.

Calif.: Claremont, Harris, Humboldt Co., 29 June; Wash.: Ains-
worth. Camp LTmatilla, Yakima River, Wenas Valley, June, July,

1882.

Anoplius.

1. Claw with tooth sloping toward tip of the claw, third joint of antennae very

long; third submarginal cell not petiolate selkirkensis.

Claw with an erect tooth, smaller 2

2. Third submarginal cell petiolate; abdomen elongate compadus.

Third cell not petiolate 3

3. Fourth joint of hind tarsus nearly'as long as the fifth, latter much shorter

than the third . tarsatus.

Fourth joint of hind tarsus plainly shorter than fifth, which is subequal
to the third luduosus.

1. Third submarginal cell petiolate, the abdomen more elongate, .compadus.
Third cell not petiolate 2

2. The ventral plate with the carina of even height throughout or nearly so.

selkirkensis.

The ventral plate with the carina very strong on basal part, and then

suddenly sloping behind luduosus.

234 bulletin: museum of comparative zoology.

Anoplius luctuosus (Cresson).

B. Col.: Revelstoke, Selkirk Mts., 30 June, Carbonate, Columbia
River, 7-12 July; Oregon: Corvallis; Calif.: Ingleside, 25 August,
Ukiah, Mendocino Co., 30 September; Wash.: Wenas Valley, Yakima
River, July, 1882.

Anoplius compactus (Provancher).

B. Col.: Glacier, Downie Creek, Selkirk Mts., 9 August; Calif.:
Sequoia National Park, 21-25 July.

Anoplius selkirkensis, sp. nov.

Type.— Cornell Univ. Coll. Paratypc— M. C. Z. 10,398.
B. Col: Downie Creek, Selkirk Mts., 14 August (Bradley).

9 Black; densely long hairy; clypeus broad, truncate in front; no hair
under first joint of antennae,' third joint very long, with the second fully equal
to vertex width; a groove above the antennae; hind ocelli closer to each other
than to the eyes; vertex slightly convex from in front; hind margin of pro-
notum angulate; metanotum with a median groove; abdomen not depressed,
venter and tip hairy; legs slender, spines short, those at tip of hind tibia not
one half diameter of the joint; inner spur of hind tibia a little more than one
half of the basitarsus; claws long, with a large tooth directed toward tip of
claw (not at right angles as in most species); last joint of hind tarsus with
stout spines beneath, wings dark, darker toward tips, the basal vein nearly
interstitial with the transverse, marginal cell hardly its length from wing-tip,
second and third submarginal cells subequal, the latter narrowed above, the
recurrents end beyond middle of the cells.

Length 9 mm.

The male is much smaller, more slender, more sericeous on face and
thorax, and without ventral tufts of hair.

Distinct from other western species by longer third antennal joint,
and the sloping tooth of the claws.

Anoplius tarsatus, sp. nov.

r^/i^e.— Cornell Univ. Coll. Paratypc.— M. C. Z. 10,399.
Calif.: Sherwood, Mendocino Co., 1 July, Sugar Pine, Madera Co.,
4,300-5,000 ft., August (Bradley).

banks: psammocharidae. 235

9 Blue-black; rather densely clothed with long hair, none below first
joint of antennae; clypeus broad, truncate in front; groove above antennae,
second and third antenna! joints together fully equal vertex width; vertex,
from in front, nearly straight across; hind ocelli rather nearer each other than
to the eyes; pronotum angulate behind; metanotum without distinct groove;
abdomen hairy on basal segment, at tip and beneath; legs long, spines moder-
ately stout, those near tip of hind tibia fully one half the diameter of the joint,
inner spur of hind tibia about three fifths of basitarsus; in the hind tarsus the
fourth joint is nearly as long as the fifth, and the third is much longer than the
fifth. Wings violaceous black, basal vein plainly before the transverse,
marginal cell slightly more than its length from wing-tip, second and third
submarginals subequal, the third much narrowed above, the second recurrent
bowed outward, reaching the third submarginal cell just beyond middle.

The claws are as usual in the genus, a small erect tooth near middle.

Length 10 mm.

In appearance this is much like A. fulgidus, but the fourth joint of
hind tarsus is much shorter in that species, the tooth on claw is nearer
to the tip of claw, and the basal vein is nearly interstitial with the
transverse.

POMPILOIDES.

1. The second discoidal cell hardly longer than broad; small species 4

The second discoidal cell plainly longer than broad 2

2. Third submarginal cell broad above elsinore.

Third submarginal cell petiolate or nearly so 3

3. Female venter hairy; cf inner spur hind tibia three fourths of the basi-

tarsus; in genitalia the lateral lobes of basal piece reach much beyond

the cleft median lobe dystera.

Female venter smooth; cf inner spur hind tibia about two thirds of the
basitarsus; in genitalia the lateral lobes of basal piece scarcely reach
beyond the emarginate median lobe estellina.

4. Abdomen wholly black; second submarginal cell subtriangular . .angularis.
Abdomen with red spots near base; second submarginal cell trapezoidal.

hageni.

POMPILOIDES HAGENI, Sp. nOV.

Type— M. C. Z. 10,400.

Mont.: AYeeksville, 2 August, 1882; Wash.: Camp Umatilla,
26 June, 1882.

236 bulletin: museum of comparative zoology.

9 Black; abdomen with rufous above on apical part of the second seg-
ment and basal part of the third segment; wings brown; tips of the antennae
and the tarsi brown. General shape of P. marginatus. Third joint of an-
tennae elongate, much longer than fourth. Vertex, seen from in front, slightly
convex; hind ocelli about as close to eyes as to each other. Pronotum angu-
late behind. Legs rather strongly spined; inner spur of hind tibia about two
thirds of basitarsus. In the wings the marginal cell is nearly twice its length
from apex of wing; the second submarginal cell trapezoidal, with sides of
equal length, the third subtriangular, receiving the second recurrent vein
beyond the middle.

Length 9 mm.

Pompiloides clystera Banks.

Calif.: Santa Cruz Island, 27 July, Lemoncove, Tulare Co., 9-11
July, Sugar Pine, Madera Co., 24-31 August, Jacintos Barranca,
near Coalinga, Fresno Co., 4 June, Sisson, 14 August, Coyote Creek,
Tulare Co., 26 June (Stanford Univ. Coll.).

Pompiloides estellina Banks.
Calif.: San Diego Co., 24 May (E. P. Van Duzee).

Pompiloides angularis Banks.

Calif.: Ukiah, Mendocino Co., 30 September, Ingleside, 25 August,
Claremont.

* Pompiloides elsinore, sp. nov.

Tt/pe.— Cornell Univ. Coll. Paratype.— M. C. Z. 10,401.
B. Col.: Carbonate, Columbia River, 2,600 feet, 7-12 July
(Bradley).

9 Black; head and venter with fine short hairs; clypeus truncate in
front, almost concave; faint line from antennae to anterior ocellus, hind ocelli
nearer to each other than to the eyes; antennae not very long, third joint
much longer than first, but barely longer than fourth, the second and third
together not as long as vertex width; vertex from in front nearly straight
across; pronotimi angulate behind; metanotum short, with a median groove;

banks: psammocharidae. 237

abdomen rather short; legs of moderate length, with many stout spines,
inner spur of hind tibia two thirds of the basitarsus, fourth joint of hind tarsus,
plainly shorter than the fifth, front tarsus with very short spines, second joint
with only one at tip. Wings pale, darker on tips, the marginal cell rather
long, second submarginal nearly quadrate, third about as large, but little
narrowed above, other veins about as usual. The male has the inner spur
of hind tibia broader than usual.

Length 8 mm.

Nannopompilus.

Bluish or greenish; body short; male shows a short basal piece to genitalia
with an angulate cleft, the ventral plate carinate padriniis.

Blackish; body more slender; the male shows a long slit with sinuous edges,
the ventral plate not carinate consimilis.

Blackish, basal part of abdomen more or less reddish rufibasis.

Nannopompilus padrinus (Viereck).

Calif. : Claremont, Ukiah, Mendocino Co., 30 September, Ingleside,
25 July.

Nannopompilus rufibasis (Banks).
Wash.: Olympia (Kincaid), agreeing with eastern specimens.

Nannopompilus consimilis (Banks).

Wash.: Wenas Valley, Yakima River, July, 1882; Oregon: The
Dalles, 23 June, 1882.

Sophropompilus.

1. Legs noticeably hairy; four comb spines on first tarsal joint 2

Legs barely hairy 3

2. Tibiae with long hairs, and a few on tarsi bradleyi.

Few if any hairs on tibiae tumifrons.

3. Pronotum slightlj^ but distinctly angulate behind subangulatus.

Pronotum broadly arcuate behind parvus.

238 bulletin: museum of comparative zoology.

1. Head tumid about ocelli, and above the antennae; femora hairy.

tumifrons.
Head not tumid at ocelli; femora not noticeably hairy 2

2. Pronotum arcuate behind ; posterior slope of metanotum silvery pubescent.

'parvus.

Pronotum angulate behind; posterior slope of metanotum not silvery

pubescent subangvlatus.

Sophropompilus bradleyi, sp. nov.

Type. — Cornell Univ. Coll.

Calif.: Giant Forest, Sequoia National Park, 6,000-7,000 ft., 21-26
July (Bradley). One 9 •

9 Bluish, legs and antennae more black; wings not very dark, not darker
at tip; body and legs densely long haired, even down on the first and second
tarsal joints, most of the hairs erect, and very long. Head rather broad,
clypeus truncate in front, margined, a distinct groove from antennae to an-
terior ocellus, hind ocelli nearer to each other than to the eyes; vertex from
in front slightly convex; antennae short and heavy, third joint as long as first;
pronotum broadly arcuate behind, metanotum short, rounded, with a median
groove; abdomen moderately elongate, hardly depressed, hairy all over,
those at the tip no longer than elsewhere; legs rather short, with numerous
stout spines; tarsal comb of four heavy curved spines on the first joint, the
last of which is longer than the next joint; inner spur of hind tibia two thirds
of basitarsus ; wings with marginal cell length from the tip, second submarginal
cell nearly square, receiving the first recurrent vein near tip, the third sub-
marginal cell short, narrowed above, receiving the second recurrent vein near
its middle; basal vein bowed, a little before the transverse, in hind wings
cubital fork interstitial with the end of the cell.

Length 10 mm.

Sophropompilus parvus (Cresson).

Calif.: Ingleside, 25 July, Sequoia National Park, 21-26 July;
B. Col.: Carbonate, Columbia River, 7-12 July; Wash.: Yakima,
and Wenas Valleys, July, 1882.

Sophropompilus tumifrons Banks.
Calif.: San Diego Co., 14 June.

banks: psammochakidae. 239

SOPHROPOMPILUS SUBANGULATUS, Sp. nOV.

Type.— Cornell Univ. Coll. Paratype.— M. C. Z. 10,402.

Colo.: Tabernash, August (Tucker Coll.); Calif.: Ingleside, 23-26
August (Bradley); B. Col.: Beaver Mouth, Selkirk Mts., 16-20
August (Bradley).

9 Black, abdomen bluish, similar to *S. ingenuus in most respects, but-
much smaller, body not very hairy, the femora show no hairs, third joint of
antennae about equal first, pronotum slightly but plainly angulate behind,
metanotum with a deep median groove; inner spur of hind tibia nearly two
thirds of basitarsus; the tarsal comb has the last spine of the first joint longer
than the next joint, and the basal two not half their length apart (in ingenuus
the last is not as long as next joint, and the basal two farther apart). Wings
dark, but paler before the basal vein, venation about as in S. ingenuus.

Length 8 to 10 mm.

Differs from S. ingenuus by longer tarsal comb, and angulate pro-
notum. It differs from S. parvus in larger size; distinctly angulate
pronotum, and in a much longer comb on female tarsi.

Agenoideus.

But one species known which occurs across the northern part of
North America.

Agenoideus humilis (Cresson).
B. Col.: Revelstoke, Selkirk Mts., 8-13 July.

Gymnochares.
Two species known, the Arizonian one extending into Califorjiia.

Gymnochares biedermanni (Banks).
Calif.: Claremont (Baker).

Arachnophroctonus.

1. Body uniformly yellowish unicolor.

Body more or less marked with brown or reddish brown . . . .ferrugineus.

240 bulletin: museum of comparative zoology.

Arachnophroctonus unicolor Viereck.

Calif.: Lenioncove, Tulare Co., 9-11 July, Dulzura, San Diego Co.,
15 August, Three Rivers, Tulare Co.; Wash.: Wenas Valley, 6 July,

1882.

Arachnophroctonus ferrugineus (Cresson).
Calif.: Claremont, National City, 15 May.

Aporinellus.

1 . Legs partly reddish calif ornicus.

Legs all black 2

2. Abdomen (except last two segments) wholly sericeous completus.

Abdomen black with sericeous bands or spots 3

3. Abdomen with complete sUvery bands across apex of each segment above.

apicatus.
Abdomen without complete bands 4

4. Pronotum and pleura noticeably marked with silvery; wings dark at tips.

intermedins.

Pronotum and pleura not noticeably marked with sericeous; wings nearly

uniformly blackish medianus.

Aporinellus californicus Rohwer.
Calif.: Alameda Co.

Aporinellus completus Banks.
Wash.: Yakima River, Kittitas Valley, June, July, 1882.

Aporinellus apicatus Banks.

Calif.: National City, 15 May, Felton, Santa Cruz Mts., 20-25
May, Berkeley, 16 September.

Aporinellus medianus Banks.

Calif.: Ramona, 15 August, Los Angeles, May, Sequoia National
Park, 6 August, Lemoncove, Tulare Co., 9 July, El Cajon.

banks: psammocharidae. 241

x\PORINELLUS INTERMEDIUS, Sp. nOV.

Type.— Cornell Univ. Coll. Paratype.— M. C. Z. 10,403.

Calif.: Clareniont (Baker), Owen's River, 5 August (Kennedy).

9 Black; marked with sericeous as follows : — most of face, back of head,
front of coxae, outer sides of femora and tibiae, pleura, part of the pronotum,
its posterior margin very strongly so, a mark on sides of the scutellum, at apex
of metanotum, and large subtriangular median spots at apex of first, second,
and third abdominal segments above. Wing not very dark, but beyond the
stigma it is black. Body rather heavy, but vertex not as broad as in laticeps;
antennae longer than in that species, the second plus third joints equal vertex
width; pronotum arcuate behind; inner spur of hind tibia a little more than
one half the basi tarsus; two or three submarginals, the third being extremely
small, if present, second recurrent vein received near apex of second or at
middle of third cell.

Length 7 mm.

Differs from A. laticeps in having the marks on abdomen triangular
(instead of complete bands), and in longer antennae; it differs from

A. medianus by the wings being dark only at tip, the sericeous marks
on thorax, and in having much longer spines in the comb of front legs.

Episyron californicus Banks.

Calif. : Ingleside, 25 August, Fresno, May 5, Three Rivers, Tulare
Co., 16 July, Felton, Santa Cruz Mts., 15-19 May, Claremont;

B. Col.: Revelstoke, Selkirk Mts., 4-6 July (Bradley).

RiDESTUs STRIATULUS Banks.
"Calif." (Cornell Univ. Coll.).

PSORTHASPIS PLANATUS (Fox).

Calif.: Laguna Beach, Lemoncove, Tulare Co., 16 May, San Diego,
14 June.

Planiceps.

1. Legs not noticeably hairy, metanotum not hairy luxus.

At least front femora and tibiae hairy 2

2. All femora and tibiae plainly hairy hirsutus.

Only front femora and tibiae hairy assimilis.

242 bulletin: museum of comparative zoology.

Planiceps luxus Banks.

Calif.: Washington (Kincaid), Claremont, Ingleside, iVugust,
National City, 15 May, Santa Clara Co., May; Wash. : Yakima Valley,
16 July, 1882.

Planiceps hirsutus Banks.
Calif. : Mountains near Claremont, Pasadena, Ingleside, 26 May.

Planiceps assimilis Banks.

B. Col.: Penicton, 11 August, Peachland, 24 August; Calif.:
Samoa Beach, Humboldt Co., 28 June.

Ageniella.

1. Body mostly yellowish, legs wholly so 2

Body mostly black (at least head and thorax) 3

2. Wings uniformly deep black; larger species coronata.

Wings paler, darker on tips and over basal vein; smaller species. blaisdelli.

3. Third submarginal cell as high as long 4

Third submarginal cell plainly longer than high 5

4. Basal segment of abdomen slender, fully three times as long as wide at

tip, slightly margined below suhaequalis.

Basal segment of abdomen not twice as long as broad at tip ... . aequalis.

5. Basal segment of abdomen about twice as long as broad at tip; wings

nearly uniformly brown; apical segment pale euphorbiae.

Basal segment plainly more than twice as long as broad at tip, broadly
margined below; wings darker at tips . . .Q

6. Abdomen basally red; legs mostly pale; mid and front spurs white.

partita.
Abdomen black, apical segment white; mid spurs black, legs mostly black.

praestans.

Ageniella coronata, sp. nov.

Type.— Cornell Univ. Coll. Paratype.— M. C. Z. 10,404.
Calif. : Mts. near Claremont (Baker), San Buenaventura, 18 August,
Santa Rosa.

banks: psammocharidae. 243

Yellowish red as in A. blaisdelli; a little black between ocelli, abdomen
sometimes darkened (discolored) above near tip; wings violaceous black,
uniformly dark throughout in both pairs. Similar to A. blaisdelli in structure;
the pronotum behind rather angulate; legs weakly spinose; distinguished by
its much larger size and uniformly darker wings.

Length 14 mm.

Ageniella blaisdelli (Fox).

Calif.: Lemoncove, Tulare Co., 9-11 July, Lompoc, 9 September,
Mesa Grande, Russian River, 30 September, Glenwood, 27 Miay,
Sugar Pine, Madera Co., 21: -31 August.

Ageniella euphorbiae (Viereck).
Cjalif. : Described from San Pedro (T. D. A. Cockerell).

Ageniella subaequalis, sp. nov.

Type. — Cornell Univ. Coll.

Calif.: Mts. near Claremont (Baker).

c? Black; tips of mandibles reddish; wings faintly fumose, hardly darker
at tip; last segment of abdomen brownish above. Body very slender; face
narrowed below, silvery sericeous on the lower part, lateral ocelli a little closer
to each other than to the eyes, last joint of the antennae strongly compressed;
hind margin of pronotum deeply emarginate behind, pleura and hind part of
the metanotum silvery; abdomen very slender, first segment about three
times as long as broad at tip, petiolate, slightly margined below toward tip.
Spurs rather paler than legs, inner spur of hind tibia a little more than one
half of the basitarsus. Wings rather short, marginal cell fully its length from
wing-tip, third submarginal plainly higher than long, only slightly narrowed
above, basal vein bowed, interstitial with the transverse, second recurrent
reaches third submarginal cell beyond the middle.

Length 4 mm.

Ageniella aequalis, sp. nov.

Type. — Cornell Univ. Coll.

B. Col.: Revelstoke, Selkirk Mts., 1 July (Bradley).

244 bulletin: museum of comparative zoology.

cf Black; front tibiae and tarsi brown, spurs brown; tips of mandibles
pale; wings uniformly light fumose, not darker on tips. Body slender;
face rather broad, nearly as broad below as above, clypeus truncate, lateral
ocelli much nearer to each other than to the eyes; posterior margin of the
pronotum angulate, metanotum not strongly sericeous; abdomen broad,
sessile, first segment hardly one and a half times longer than broad at tip,
apical segment with short fine hairs, the inner spur of hind tibia a little more
than one half of the basitarsus; third submarginal cell nearly square, but a
little higher than long, and scarcely narrowed above, receiving the second
recurrent at middle, basal vein bowed, much before the transverse.

Length 5 mm.

Ageniella praestans Banks.
Calif.: Muir Woods, 30 August, Claremont, June.

Ageniella partita, sp. nov.

Type.— Cornell Univ. Coll. Pamtypc— M. C. Z. 10,405.
Calif.: Brawley, Imperial Co., 9 August (Bradley).

cf Black, basal part of the abdomen mostly reddish or yellowish, apical
segment white above, scape beneath, narrow lower margin of clypeus, and the
tips of mandibles pale; legs mostly pale, front and mid spurs white; wings
pale, darker on tips. Slender; face slightly narrowed below, silvery sericeous,
lateral ocelli rather nearer to eyes than to each other, hind border of pronotum
broadly arcuate, thorax silvery sericeous, especially pronotum, spot before
wings, scutellum, metanotum and two oblique stripes on the pleura. Abdomen
slender, basal segment about three times as long as broad at tip, broadly
margined below, with a dark stripe above. Front legs, and femora and tibiae
of the other pairs pale, the coxae may also be pale below; inner spur of hind
tibiae a little more than one half of the basitarsus. Marginal cell not its length
from the wing-tip, third submarginal cell much longer than high, narrowed
above, receiving the second recurrent vein near middle, basal vein is inter-
stitial witli the transverse.

Length 4.5 to 5 mm.

A female from Sugar Pine, Madera Co., Calif, may belong here;
it is black with a red abdomen, black legs; but the general structure
is similar.

banks: psammocharidae. , 245

PSEUDAGENIA METALLICA Banks.

Calif.: Ramona, 15 August, Santa Clara Co., May, Blue Lake,
Humboldt Co., 20-27 June, Mts. near Claremont (Baker), Claremont;
B. Col.: Revelstoke, Selkirk Mts., July; Wash.: Wenas Valley,
Yakima River, July, 1882.

Priocnemis.

1. Wholly black 2

Abdomen more or less reddish 3

2. Metanotum transversely wrinkled; clypeal margin concave. . . .hesperus.
Metanotum not striate or very minutely so; clypeal margin truncate.

pompilus.

3. Head, thorax, coxae, and femora very hairy, larger species. . . .comparatus.
Thorax, coxae, and femora not noticeably hairy . .4

4. Dark cloud in wing; clypeus all black; inner spur hind tibia one half of

basitarsus alienatus.

No dark cloud in wing; clypeal margin pale; iimer spur of hind tibia not
one half of the basitarsus placitus.

Priocnemis comparatus (Smith).

Washington (Kincaid); Calif.: Blue Lake, Humboldt Co., 20-26
June, Mendocino, San Jose.

Priocnemis alienatus (Smith).

B. Col: Revelstoke, Selkirk Mts., 4-6 July.

Eastern specimens do not differ from those from Revelstoke.

' Priocnemis placitus (Banks).

Calif.: Lemoncove, Tulare Co., 7-11 July.

Priocnemis pompilus Cresson.

Washington (Kincaid); Calif.: Felton, Santa Cruz Mts., 15-16
May.

246 bulletin: museum of comparative zoology.

Priocnemis HESPERUS Banks.
Calif.: Stanford Univ., 3 September.

Cryptocheilus.

1 . Wings more or less yellow or reddish 2

Wings, and antennae black atratus.

2. Antennae yellowish 3

Antennae black 4

3. Second recurrent strongly bent; in hind wing the cubitus arises before

the end of the cell flammipennis.

Second recurrent only slightly curved; in hind wing the cubitus arises
beyond the end of the cell pallidipennis.

4. Abdomen hairy above, femora also very hairy 5

Abdomen not hairy above except at tip, femora not noticeably hairy.

terminatus.

5. Metanotum wrinkled on basal part rugosus.

Metanotum not wrinkled inaequalis.

Cryptocheilus terminatus (Say).

Calif.: Claremont, Los Gatos Divide to Mt. Diabolo, 6-8 June.
Coalinga, 9 June; Wash.: Yakima River, 30 June, 1882.

Cryptocheilus flammipennis (Say).
Calif.: Lemoncove, Tulare Co., 7-11 June.

Cryptocheilus inaequalis Banks.
Wash.: (Kincaid), Camp Umatilla, 27 June, 1882.

Cryptocheilus pallidipennis (Banks).
Calif.: Brawley, Imperial Co., 9 August.

Cryptocheilus rugosus Banks.
Idaho: Lapwai, 4 August; Wash.: Wawawai, June.

banks: psammocharidae. 247

Cryptocheilus atratus, sp. nov.

Type.— Cornell Univ. Coll. Paratypc— M. C. Z. 10,406.

Calif.: Glenwood, 27 May, Lemoncove, Tulare Co., 9-11 July,
Harris, Humboldt Co., 29 June, Felton, Santa Cruz Mts., 20-25 May
(Bradley).

9 Deep black, the tarsi more brown, wings violaceous. Body with short
hairs, longer on the vertex, tip, and venter of abdomen. Face as broad below
as above, clypeus slightly concave below, lateral ocelli very much nearer to
each other than to the eyes, vertex, from in front, nearly straight across, third
antennal joint one and a half times longer than first, one and a fourth longer
than the fourth joint, last joint very slender. Pronotum slightly angulate
behind; metanotum transversely striate, most noticeable on the sides. Ab-
domen slightly depressed, hair at tip rather yellowish brown. Legs slender,
mid and hind tibiae with numerous short, but stout spines, inner spur of hind
tibia about two fifths of t^e basitarsus. Wings hardly reaching beyond
abdomen; marginal cell nearly its length from wing-tip, second submarginal
cell little longer than broad, receiving the first recurrent vein beyond middle,
the third submarginal longer than the second, not extending beyond marginal,
receiving the second recurrent vein near middle, the latter evenly but not
strongly curved, basal vein before the transverse.

Male is much more slender, the pronotum is plainly much longer, suggesting
a Pedinaspis, the metanotum is not distinctly striate; the lateral parts of the
genitalia are densely black fringed on the outer side.

Length 8 to 14 mm.

Related to C. idoneus from North Carolina but latter has the
metanotum not plainly striate, less violaceous wings, and slightly
different venation.

Mysnimia.

1. Second recurrent vein nearly straight across .hespcrina.

2. Second recurrent vein much curved ustulata.

Mygnimia ustulata (Dahlbon).
Utah: Salt Lake Co., 30 May; Arizona.

Mygnimia hesperina Banks.
Calif.: San Diego, Stanford Univ.

248 bulletin: museum of comparative zoology.

Pepsis.

1. Antennae more or less reddish or yellowish , mildei.

Antennae all black 2

2. Wings reddish cinnabarina.

Wings yellowish formosa.

Pepsis Formosa (Say).
Calif.: San Emigdio Canon, Kern Co., cf 9 •

Pepsis cinnabarina Lucas.
Calif.: Claremont, Los Angeles, Kern Co.

Pepsis mildei Stal.
Calif. : Claremont, Pasadena, San Jose, San Luis Obispo.

Ceratopales.

1. Wings black nigripes.

Wings hyaline 2

2. First abdominal segment almost wholly yellow above stretchi.

First abdominal segment black, with two transverse yellow spots, some-
times connected fraterna.

Ceratopales nigripes (Cresson).
Recorded from Washington.

Ceratopales fraterna (Smith).

Oreg. : The Dalles, June, 1882; Calif.: Kern Lake, 27 July, Sonoma
Co. ; Wash. : Yakima River, July, 1882, Wenas Valley, July, 1882.

Ceratopales stretchi (Fox).
Described from "California."

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.
Vol. LXIII. No. 6.

PACIFIC COAST POLYCHAETA COLLECTED BY
ALEXANDER AGASSIZ.

By Ralph V. Chamberlin.

With Three Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM

November, 1919.

No. 6.^ Pacific Coast Polychacta collected by Alexander Agassiz.

By Ralph V. Chamberlin.

In 1859 and the early part of 1860 a considerable collection of
Pacific Coast polychaetes was made by Alexander Agassiz in the
intervals of his work as Aid on the U. S. Coast Survey. The greater
part of the material was secured at Mendocino and Crescent City,
California, which are on a section of the coast scarcely represented
in more recent collections of polychaetes. Other specimens were
secured at San Mateo, Calif, and on the Gulf of Georgia, Washington,
and a few at Panama. This collection, though of much interest and
value, has remained unidentified until the present time with the excep-
tion of a number of species of Nereis described by Ehlers in his Die
Borstenwiirmer (1864-1868). The following is a complete list of the
species in the collection at present identifiable. In addition to the
specimens collected by Mr. Agassiz himself, notes upon some material
secured by others on the Pacific Coast during or near the same period
of time are also included

Amphinomidae.
1. NoTOPYGOS maculatus (Kinbcrg).

Lirione maculata Kinberg, Ofvers. K. vet. akad. Forh., 1857, 14, p. 12.

Two large specimens from Panama, one collected by Dr. G. A,
Maack on the Darien Expedition and one by Q. F. Davis (received
M. C. Z. in 1862) are referred to Kinberg's species, the type of which
came from the same locality, on the assumption that this type is a
very young specimen. Kinberg's^ account is so meager that, were the
present specimens not from the same locality, they would scarcely
be referred with much confidence to this species. These specimens
are much larger than the type, approaching rather crinitus or splendens
in size. One consists of thirty-three and one of thirty-five segments.
The dark maculations and stripes are strongly developed. The
Philippine specimens referred by Grube (Annulata Semperiana, 1878,
p. 8, pi. 1, fig. 3) to this species are not conspecific with the Panama
form, as shown by their much simpler branchiae, different coloration,
and other features.

252 bulletin: museum of comparative zoology.

2. Eurythoe complanata (Pallas).

Aphrodita complanata Pallas, Misc. zool., 1766, p. 109, pi. 8, fig. 1926.

Numerous specimens of this widespread form from Panama (C. F.
Davis, received M. C. Z. in 1862), and Mazattan, Mexico (H. Ed-
wards, received M. C. Z. in 1873).

Polynoidae.

3. Halosydna insignis (Baird).

Lepidonotus insignis Baird, Proc. Zool. soc. London, 1863, p. 106.

Many deep-colored specimens collected by Mr. Agassiz at Crescent
City, in May, 1859. Numerous specimens labeled simply "Cali-
fornia, A. Agassiz." A specimen from Sacramento Bay, Calif., was
collected by Capt. H. Davis in April, 1859.

4. Lepidonotus caeloris Moore.

Proc. Acad. nat. sci. Philad., 1903, p. 412, pi. 23, f. 12.

Polynoe squamatus Johnson (non Linne), Proc. Calif, acad. sci. Zool., 1897,
ser. 3, 1, p. 166, pi. 7, f. 30; Treadwell, Univ. Calif, publ. Zool., 13,
p. 181.

Two specimens San Francisco, Calif. (T. G. Cary) and one Gulf of
Georgia, Wash. (A. Agassiz).

Hesperonoe, gen. nov.

Differing from Antinoe in having notopodial setae of two very different
forms, the more numerous ones much stouter than the neuropodials and a
fine, slender, capillary form. Neuropodials also of two forms, the more
slender supraaciculars of the Antinoe type with the long, fine, smooth tips,
and the subaciculars mostly stouter and with tips more approaching the
Eunoe type. Otherwise agreeing with Antinoe.

Genotype. — H. senilis, sp. nov.

chamberlin: pacific coast polychaeta. 253

5. Hesperonoe senilis, sp. nov.

Plate 1, fig. 1-4.

Differing decidedly from Antinoe macrolepida Moore, known from the Gulf
of Georgia and northward, in having the peaks of the prostomium conspicu-
ously large, pointed, and widely separated. Anterior eyes much smaller than
in that species, but little exceeding the posterior. Median tentacle slender,
tapered, not morq. than twice as long as the prostomium measured to anterior
ends of peaks; lateral tentacles very much smaller as usual, but little exceed-
the ceratophore of the median tentacle and shorter than the prostomium.
Prostomium longer than wide. Palpi very long and evenly tapered, much
exceeding the median tentacle. Tentacular cirri similar in form to the palpi
but shorter and proximally more slender; the inferior one thicker than the
dorsal. Body ventrally widest near somite XII, from where it narrows
evenly and moderately caudad, proportionately much wider than in macro-
lepida, the ventral width much exceeding the length of parapodia exclusive
of cirri and setae. No distinct ventral groove. Nephridial tubercles begin-
ing on VII, slender. Elytra overlapping moderately medially and cephalo-
caudally. Those of first pair circular, the others elliptical; second, third,
and fourth pairs broadly and shallowly incurved on anteroectal side, the first
of these the most strongly so. Elytra thin, marginally transparent, elsewhere
translucent; surface, excepting a narrower border, covered with low, light
colored tubercles, each tubercle bearing a dark spinous point; the tubercles
not crowded, becoming much smaller on anteroectal part of scale, the posterior
ones commonly more or less laterally compressed; edges smooth excepting
on external side where there are fine short cilia. Notocirri much exceeding
the setae, slenderly and evenly tapered with no sub terminal enlargement;
cirrophore large. Neurocirri attached distad of middle, slenderly subulate,
obviously shorter than the shortest setae. Each ramus of parapodium sup-
ported by a single, stout, evenly tapering aciculum, the notopodial stouter
than the neuropodial. Notopodial setae numerous but fewer than the neuro-
podials; the prevalent type shorter but much stouter than the stoutest neuro-
podials, a few most dorsal ones very short. In addition in the ventral part
of the fascicle a number of very fine but longer capillary setae with widely
separated teeth along one edge. The supracicular neuropodials are of the
Antinoe type, being slender with exceptionally long and fine smooth tips. Of
the much more numerous subacicular setae the majority are coarser than the
supraoculars with tips notably shorter, approaching more the Eunoe type.
Below these a group of much finer setae with shorter heads set at a greater
angle to the shaft. The general color of parapodia and body at present light
brown, the body above and below with a paler median longitudinal band;
elytra greyish or colorless. Last few segments of body of type missing.

254 bulletin: museum of comparative zoology.

Length to caudal end of twelfth elytra 23 mm.; width to tips of
setae, 9.5 mm. ; to outer edges of elytra, 7.8 mm. ; to bases of parapodia
ventrally, 3.6 mm.

Locality. — Calif.: San Francisco (A. Agassiz).

Type.— M. C. Z. 179.

6. Harmothoe imbricata (Linne).

Aphrodita imbricata Linne, Syst. nat., ed. 12, 1767, 1, p. 1084.

One specimen from San Francisco (T. G. Cary) and one from
Crescent City (A. Agassiz). This is a common species on the Pacific
Coast from Alaska; to San Diego.

Aphroditidae.

7. x\pHR0DiTA armifera Moore.

Proc. Acad. nat. sci. Philad., 1910, p. 371, pi. 31, fig. 65, 66, pi. 32, fig. 67-75-

Two specimens of this strongly characterized species were dredged
in 22 fathoms of water five miles south of Santa Barbara.

8. Aphrodita leioseta, sp. nov.

This form is close to A. castanea Moore, from which it is separated chiefly
because of differences in the setae. The upper neurosetae are black, the
median ones black or dark brown and the ventrals brown of a usually lighter
shade. The neurosetae of the ventral series are from two to five in number,
three and four being most common in the middle region of body, as against
from she to eight in castanea. Three in the middle series and two in the upper
as usual. The neurosetae differ from those of castanea in being all wholly
smooth, with no trace of hairiness at tips, and in being distally more strongly
curved, the curvature weakly sigmoidal, with tip less acute, that of the dorsals
in particular narrowly rounded. The notosetae are proximally dark or
reddish brown but in middle region of length become light and distally color-
less. They have the long, soft, slender distal ends characteristic of the
japonica group though these do not seem to be normally at all hooked. As a
whole they are strongly curved and are largely concealed by the felt, the distal
portion in all cases normally buried in this. The prostomium is long, some-
what inversely pyriform, being much narrowed proximad; evenly rounded

chamberlin: pacific coast polychaeta. 255

anteriorly, without distinct ocular peduncles. Eyes on dorsal surface, the
anterior farther apart than the posterior from which it is separated by about
its radius. Median tentacle very short, clavate. Facial ridge remarkably
long, much exceeding the prostomium in length, acutely narrowed anteriorly.
Elytra large, overlapping in middle line. Dorsal felt abundant and dense,
greyish brown, normally involving and dorsally concealing the neurosetae in
dorsal view. Number of somites, thirty-seven.

Length of body proper, 33 mm. ; length over all, 38 mm. Greatest
width to bases of parapodia, 12.5 mm. Width to tips of neurosetae,
24 mm.

Locality. — Calif.: Mendocino (A. x\gassiz).

Type.— M. C. Z. 14.

Nepthydidae.
9. Nepthys caeca (Fabricius).

Nereis caeca Fabr., Fauna Groenlandica, 1780, p. 304.

Two specimens from the Gulf of Georgia, Wash. (A. Agassiz).

Nereidae.
10. Nereis californica Ehlers.

Borstenwurmer, 1868, p. 553, pi. 23, fig. 2.

Ehlers founded this species on a single incomplete specimen taken
by Mr. Agassiz at Mendocinch In the collection I find several
additional specimens taken at this locality at the same time.

Type.— M. C. Z. 40.

11. Nereis procera Ehlers.
Borstenwiirmer, 1868, p. 557, pi. 23, fig. 2.

Taken in Gulf of Georgia, W^ash.
Type.— M. C. Z. 155.

256 bulletin: museum of comparative zoology.

12. Nereis brandti Malmgren.

Ofvers. K. vet. akad. Forh., 1865, p. 183.

Sacramento Bay (Capt. H. Davis, April, 1859), San Mateo, Calif.,
and Gulf of Georgia, Wash. (A. Agassiz).

13. Nereis vexillosa Grube.

Middendorf's Reise nord. u. ost. Siber. Zool., 1851, 2, th. 1, p. 4, pi. 2, fig. 4,
5,6.

Mendocino, Crescent City, and San Mateo, Calif., and at the
Gulf of Georgia, Wash. (A. Agassiz).

14. Nereis mendocinana, sp. nov.
Plate 1, fig. 5.

'b*

In this species the notopodia are simple, none at all enlarged into a large
lamellar form bearing the cirrus such as occurs in brandti and vexillosa. The
two lobes of the notopodia are equal throughout. In a considerable number
of the anterior notopodia the lobes are thick and conical; but caudad they
become more slender and thin in the anterocaudal direction, in the posterior
region appearing as simple, short triangular processes. Notocirrus attached
at base of dorsal lobe, long and slender or filiform. No postsetal lobe in the
anterior parapodia but caudad one becomes evident as a low, but broad, dis-
tally straight or slightly convex tip. Lower neuropodial lobe conical, nearly
the same size as the notopodial lobes, becoming smaller and more slender
caudad. Neurocirri arising from swelling at base of neuropodia, slender, sur-
passing neuropodial lobe. Anterior region of prostomium triangular with
anterior end narrowly truncate and bearing the tentacles which are contiguous,
subulate, and shorter than the distance between eyes. Eyes large, the poste-
rior eye on each side in line with the anterior or nearly so, smaller. Styles
of cirri short, articulated. In the proboscis I is unarmed. Each II bears nine
or ten teeth in a double oblique line. V is unarmed. Each VI bears a few
denticles in a central area. On each IV there are ten teeth, of which eight
or nine form a curved line while one or two lie on the concave side of this.
On the ventral side across VII and VIII run several series of denticles of which
those of the most anterior series are fewer, more widely separated and much

chamberlin: pacific coast polychaeta. 257

larger than those of the more posterior series. Each maxilla with six teeth.
Number of segments, seventy.

Length, 33 mm. ; greatest width, exclusive of parapodia, 3 mm.
Locality. — CaHf. : Mendocino (A, Agassiz).
Type.— M. C. Z. 2,129.

15. Nereis agassizi Ehlers.

Borstenwiirmer, 1868, p. 542, pi. 23, fig. 1.

Mendocino, Calif. (A. Agassiz).
Type.— M. C. Z. 149.

Leodicidae.

16. Leodice valens, sp. nov.

Plate 1, fig. 6-8.

In the type of this species the branchiae begin on somite V and end on
somite LXXX on the right side and LXXVIII on the left. The first and
second branchiae are simple filaments; the third are bifid with the two fila-
ments equal; the fourth gills, like the succeeding compound ones, unilaterally
pectinate, the branches five in number, each forming an acute angle with
base. Caudad of this the number of branches increases to a maximum of
ten, the branchiae also longer, when laid against dorsum attaining or extending
a little beyond the mid-dorsal line. Farther back they again decrease in
length and in number of filaments, and the last seven to nine are simple fila-
ments like those of the first two pairs. Anterior border of prostomium emar-
ginate or incised at the middle between bases of palpi. Anterior lateral ten-
tacle on each side inserted cephaloectad of the posterior and directly in front
of the eye. Posterior paired tentacle a little farther forward than the median,
the bases of all forming a semicircle. Ceratophores extremely low. Style
of median tentacle evenly tapered, not constricted between rings which are
cylindrical, long, and about twelve in number. Nuchal cirri nearly reaching
anterior edge of peristomium; rings eight or nine, not at all moniliform.
Mandibles with anterior plates white, anterior margin of each with two incurv-
ings and three rounded teeth or broad crenulations. Left maxilla II with six
teeth; the right with eight of which the most caudal are small; the unpaired
or mesal left plate with eight (or nine) teeth. Left maxilla III with six teeth,
the right with eight. Notocirri mostly flattened subligulate; the anterior

258 * bulletin: museum of comparative zoology.

ones much larger than the posterior; most not showing any distinct segmenta-
tion, but some of the anterior ones more or less divided into a long basal
segment and usually three smaller distal ones. Neurocirri all thick, short,
subconical, and much exceeded by the setae. Shaft of composite setae strongly
enlarged distally. Appendage bidentate, the upper tooth erect, the smaller
lower one forming a very obtuse angle with it, and nearly at right angles to the
axis of the appendage; guard narrow; length of appendage less than once
and a half the diameter of the exposed portion of shaft. General color light
brown, branchiae paler; no distinct markings. Anal cirri long, slenderly
tapered. Total number of segments one hundred and eighty-one.

Length, 140 mm.; greatest width, 7 mm.
Locality. — CaHf. : Mendocino (A. Agassiz).
Type — M. C. Z. 120. Paratype.— 121.

LUMBRINEREIDAE.
17. LUMBRINEREISZONATA JohnSOH.

Proc. Boston soc. nat. hist., 1901, 29, p. 408, pi. 9, fig. 93-100.

Several small specimens referred to this species differ from typical
larger specimens in wholly lacking brown annulations, being uniform
yellow of a slight ferruginous tinge. The crochets and setae are
relatively longer than in the type but agree in distribution. One
immature specimen from Crescent City and several from Mendocino
(A. Agassiz).

18. Arabella munda, sp. nov.

The general color light reddish brown with a darker longitudinal stripe,
which is not sharply limited, along each side of the dorsum just above the
parapodia. Body moderately narrowed from the middle cephalad but much
more strongly so caudad, the posterior end being pointed.

This species in some features much suggests A. attenuata Treadwell but
appears obviously distinct in the structure of the maxillae. These are black
throughout. Right maxilla I with seven (six) large teeth, the fang at distal
end bent mesad almost at right angle. Maxillae II strongly asymmetrical as
in A. mimetica Chamberlin. The right plate long and extending proximad
down the mesal side of the dental line of I to base of latter; bearing nine or

chamberlin: pacific coast polychaeta. 259

ten teeth and at anterior end a large curved fang well separated from teeth
by a smooth edge. Right maxilla II much shorter, ending posteriorly at
anterior end of dental series of I; toothed all the way to anterior fang, there
being seven rather blunt teeth excepting the one adjacent to the fang which is
much smaller and more acute than the others. The left maxilla II bears six
teeth inclusive of the terminal fang, the tooth at the base of which is much
smaller than the others, the third tooth from the proximal end largest. Right
maxilla III with four teeth, the fang more widely separated because of the
obliteration, or nearly so, of the minor tooth adjacent to it in the left plate.
Right maxilla IV with four small even teeth. Plate V reduced to a small
hook as usual.

Prostomium broadly ovate, more rounded anteriorly than represented for
A. aUenuata; clearly longer than the first two somites together. Eyes in a
transverse row across base as usual; the median ones more sharply defined
than the lateral ones but much smaller, the reverse of the relation in aUenuata.

First achaetous segment much longer than the second.

Posterior lobe of parapodia large, projecting caudoectad and also curving
more or less dorsad, distally bluntly rounded, always much exceeded by the
setae. Setae in middle region of body mostly from five to eight in number
in each parapodium. Upper ones with long shafts, strongly bent at beginning
of limbate portion, the distal curve gentle. Number of segments in type
three hundred and fourteen.

Length, 95 mm.; greatest width, 2.2 mm.
Locality. — CaHf . : Crescent City (A. Agassiz).
Tijpe.— M. C. Z. 211.

19. BiBORiN ECBOLA Chamberlin.

Pomona College journ. zool. ent., 1919, 11, p. 13.

One specimen taken between tide marks on San Miguel Island
by W. G. W. Harford. It is a larger specimen than the type, being
135 mm. long and having a maximum diameter of 2.5 mm. though
composed of fewer somites, — about two hundred and forty-five as
against two hundred and seventy-seven. It is brown in color with
superficial iridescence. The prostomium is less pointed than in the
type, anteriorly more broadly rounded, though this seems in part due
to shrinkage from preservation. The posterior process of parapodium
is longer in general proportionately to the basal part of parapodium
and to the setae.

260 bulletin: museum of comparative zoology.

Glyceridae.

20. Glycera robusta Ehlers.

Borstenwiirmer, 1868, p. 656, pi. 24, fig. 31, 32.

Ehlers based this species upon four specimens forming part of this
collection and coming from San Francisco and Mendocino. These
types cannot now be found in the collections of the M. C. Z. to which
Ehlers states they belong. In the Museum, however, are some finely
preserved topotypes from Mendocino (A. Agassiz). Other specimens
of the species are simply labeled "California, Capt. Brown." I have
found this species not uncommon on Monterey Bay.

Ariciidae.

21. Nainereis longa Moore.

Proc. Acad. nat. sci. Philad., 1909, p. 264.

Several specimens taken at Crescent City (A. Agassiz).

22. Nainereis nannobranchia, sp. nov.

Plate 2, fig. 10; Plate 3, fig. 1.

This form differs from the others known from the coast of California in
having the branchiae begin much farther caudad, the first ones appearing on
XX to XXIII, and in the marked reduction of the branchiae in the posterior
region. The first ones are small and tubercle-like. The others soon increase
caudad to stout conical forms, which curve mesad but with those of opposite
sides always well separated by a wide mid-dorsal space, and then, in posterior
region, again becoming more slender and much shorter. In structure and
arrangement of setae most resembling A'', hespera Chamberlin; but, aside
from the very different branchiae, readily distinguishable from that species in
the different form of the prostomium, this lacking the anterior median emargi-
nation, being simply rounded and as a whole semicircular in outline with
anterior end, however, a little narrowed. Postsetal lobe of anterior notopodia
larger, subconical, becoming smaller caudad. Postsetal lobe of thoracic
neuropodia vertically elongate, but low with edge broadly convex, decreasing

chamberlin: pacific coast polychaeta. 261

in height caudad; replaced in abdominal region by a small conical process.
Thoracic neuropodials in three subvertical series in addition to a ventrocaudal
fascicle of longer, capillary cross-striate setae. The coarser setae of the
posterior series are continuously narrowed distad with apex narrowly rounded ;
each abruptly, strongly bent, nearly geniculate, with terminal region long;
without serration or cross-ridging on the convex side, but some weak cross
wTinkles indicated on the angle on the concave side. Setae of the other two
series proximally stout, narrowing abruptly into the usual long, slender, distal
region, this with a double curve, strongly finely cross-ridged on the concave
side of proximal curve and some distance proximad of this on stouter part of
seta. Twenty-six segments in anterior region. Total number of segments
in type near two hundred and thirty-seven.

Length of type about 50 mm. ; greatest width, 3 mm.
Locality. — CaHf. : Mendocino (A. Agassiz).
Type.— M. C. Z. 2,136. Paratype.—M. C. Z. 111.

Sabellariidae.
23. Sabellaria californica Fewkes.

Bull. Essex inst., 1889, 21, p. 130, pi. 7, fig. 3, 4.
Mendocino (A. Agassiz).

24. Sabellaria nanella, sp. nov.

Plate 2, fig. 5-7.

This is a very small species readily distinguishable by the forms of the
opercular paleoli. The paleoli of the outer series have long, slender, closely
contiguous stems, normally mostly concealed, the free part expanding into
broad, colorless, or weakly golden shining blades which narrow a little distad;
the distal end finely pectinate with a process or spine toward ventral end of
series much longer and stouter than the other. On each side they number
twenty-nine or thirty and form a close semicircle spreading out horizontally,
i. e., at right angles to long axis of body. The paleoli of the inner series are
also arranged in a semicircle with convexity ectad. They are plate-like and
contiguous at base but narrow strongly and acutely distad, the tips slightly
bent ventrad so that the end appears narrowly truncate. Each blade con-
tinues just above base on ectal side into a rounded, short, and broad, process
or spur. They number about twenty on each side. The paleoli of the middle

262 bulletin: museum of comparative zoology.

series are expanded above their bases into short broad plates shaped some-
what hke the head of an adze with edge of blade ectad and narrowed end
mesad, the plates lying contiguously as a pavement or with edges more com-
monly overlapping, between bases of the other two series of paleoli. They
number about twenty-two in each series. The opercular lobes together as a
whole are long, strongly expanded distally, in a trumpet-form greatly exceed-
ing the rest of the body in diameter. The parathoracic notopodial paleoli
are elongate, thin blades with sides nearly parallel to near tip where they
expand a little clavately and then narrow abruptly to an acute apex. The
body as a whole is very slender, in all cases with a segment at anterior end of
abdomen characteristically globularly thickened.

The greatest thickness of thorax about 1.8 mm., while the width
across end of the opercular lobes is up to 1.6 mm.

Locality. — Calif.: San Francisco (A. Agassiz). Numerous speci-
mens.

Type.— M. C. Z. 2,132. Pamtypes.— l^l .C. Z. 482.

25. Idanthyrsus ornamentatus, sp. nov.

Plate 3, fig. 2-5.

General color brown. On each side of operculum at about middle of length
a large, dark, almost black, spot with a line-like dorsal prolongation to the
dorsal furrow. Also a narrow deep colored stripe below the outer series of
paleoli on each side. Achaetous caudal region dark anteriorly. Outer paleoli
yellow, the inner ones darker, bronze colored. Outer paleoli pinnately
branched and the inner ones slender and wholly smooth. Outer paleoli in a
series extending around anterior end of inner series, thirty-six in number.
Inner paleoli eleven or twelve in each series. Area outlined by the two series
on each side very narrow. Papillae below outer paleoli short, conical, well
separated, fifteen on each side of which the anterior three are longer than the
others. Two pairs of nuchal hooks present in the type. Second setigerous
somite with three cirri above setigerous papilla on each side; of these the
most dorsal is largest and corresponds in form, size, and position to the
branchiae of the succeeding somites; the two below this much shorter, stout,
and rounded. The dorsal thoracic paleoli nine or ten in number; not at all
clavately widened distad, the plates rather narrow with sides parallel to
acutely attenuated distal region, this acuminate region rather long with the
narrowing gradual and even, the species in the form of the paleoli being readily
distinguishable from /. johnstoni_ (Mcintosh) and I. armata (Ivinberg) which it
resembles in the form of the opercular paleoli. Ventral thoracic setae very
slender. Uncini elongate and slender, much as in I. regina Chamberlin but

chamberlin: pacific coast polychaeta. 263

with the margin opposite the teeth more strongly and evenly curved and the
body moderately widened and strongly rounded at end; bearing eight long,
slender teeth. Setigerous somites sixty. The achaetous appendage very
short.

Length near 50 mm. Width across thorax, near 5.5 jnm. ; width
across opercular lobes to outer ends of paleoli, 8.8 mm.; to bases of
paleoli 5.5 mm.

Locality.- — Calif.: Mendocino (A. x\gassiz).

Type.— M. C. Z. 156.

Cirratulidae.

26. AuDouiNiA spiRABRANCHUs (Moore).

Cirratulus spirabranchus Moore, Proc. Acad. nat. sci. Philad., 1904, p. 492,
pi. 38, fig. 26-29.

Numerous specimens referable to this species were taken by Mr.
Agassiz at Mendocino and Crescent City.

27. Cirratulus exuberans, sp. nov.

This species is well characterized by its prostomium. This is short and wide;
the anterior margin wide, as a whole but moderately convex, indented on each
side of the middle so as to present three large, low lobes or crenations. An-
terior, more depressed, region crossed by a single series of eyes, the series
continuous, not interrupted in mid-dorsal region, and consisting of twenty-two
eyes of which the five in the mid-dorsal region are smaller than those on each
side of them. The peristomium dorsally divided into seven or- eight short
subdivisions by transverse sulci. The special dorsal branchiae in two dense
bands on the first setigerous somite, the two bands narrowly separated in the
mid-dorsal region, the branchiae in these groups nmnerous and, like the others
very long and forming a dense tangled mass. Other branchiae present nearly
to the caudal end, about the sixteen to twenty last segments, however, appear-
ing to, lack them. They are situated unusually high on the dorsum in the
posterior region, the space between them and the notopodia on each side much
exceeding the distance between notopodia and neuropodia, but in the anterior
region they come to arise close above the notopodia. Notopodia and neuro-
podia a nearly uniform distance apart throughout length, more ventrad than

264 bulletin: museum of comparative zoology.

usual. Neuropodial crochets begin on or near segment XXV. Total number
of segments in type one hundred and fifty-five.

Length 58 mm.; width 5 mm.
Locality. — Panama.
Type.— M. C. Z. 1,285.

Terebellidae.

28. PiSTA BREVIBRANCHIA, Sp. nOV.

Plate 2, fig. 1-4.

This species much resembles P. fratrella Chamberlin, known from Laguna
Beach, Cal., which it approaches in form of the uncini of the several regions
more closely than it does P. alata Moore. It differs from both these species
in having much smaller branchiae with fewer branches and much shorter
trunks. There are two pairs, arising on II and III respectively. Of these
the posterior much exceed the anterior in size, with the right posterior largest.
The anterior branchiae with terminal branches very short, the entire organ in
length scarcely equalling that of the trunk of the right posterior. Branchiae
of each pair attached close together, trunks nearly contiguous at middle line.
Tentacles in a transverse series, attached by contracted bases; short and rather
thick but distally moderately tapered. On tentaculiferous ridge above and
on anterior surface of peristomial collar-lobes above numerous short papillae.
Peristomium with anterior edge produced into a prominent flaring collar with
ends widely separated above; deeply emarginate ventrally, leaving the lateral
portions as prominent wings. Second somite with anterior edge produced
actoss ventral surface and part way up each side, the wing laterally over-
lapped on each side by the larger one of III. The latter is scarcely evident,
obsolete, across venter, but forms a very prominent wing on each side which
rises to a level in front of the setigerous tubercle of IV and thus considerably
farther dorsad and mesad than the peristomial collar; the wings are not
united across the dorsum, the dorsal surface of III simply depressed below the
level of IV. Along the anterior edge of IV on each side also a well-developed
wing which, however, is much lower than that of III and does not extend so
far dorsad. On V and VI are similar but smaller wings which do not extend
half way to the setigerous papillae, and a smaller one on each side of VII.
The thoracic setae are distally moderately curved with the limbus on the con-
vex side broad, the other one narrow. As in fratrella the uncini show pro-
gressively reduced manubria in the anterior double rows as well as in the
single rows farther forward. In these there is a distinct rounded shoulder
besides at the base of the manubrium on the side toward the neck of the plate.

CHAMBERLIN : PACIFIC COAST POLYCHAETA. 265

In the posterior thoracic uncini the base shows a distinct angulation or shoulder
at or toward the base of the neck in place of the triangular median process
shown in alata. This shoulder is also present in the smaller abdominal un-
cini; it is more abrupt and prominent, less gentle and rounded than in fratrella
while it is quite absent in alata. The beak in the manubriate uncini is more
divergent than in fratrella.

Greatest width, 3 mm. The specimen incomplete posteriorly-
Forty-nine segments in the two pieces present.
Locality. — Calif.: Mendocino (A. x\gassiz).
Type.— M. C. Z. 502.

29. EuPOLYMNiA REGNANS Chamberlin.

Mem. M. C. Z., 1918, 48, p. 433, pi. 79, fig. 1-3.

One large and several partly grown specimens of this species were
collected by Mr. Agassiz at Panama, the type-locality, where the
species is apparently common.

30. EuPOLYMNIA CRESCENTIS, Sp. nOV.

Plate 3, fig. 6, 7.

This species differs from the widespread Indo-Pacific E. trigonostoma
Schmarda (syn. P. congruens Marenzeller) and other previously described
species in the form of the uncini. The uncini as compared with those of
trigonostoma are much more slender, evenly curved, with the beak propor-
tionately longer, reaching nearer to the end of the plate; the subrostral process
nearer the end away from the base of the beak, farther removed from tip of
beak; the basal projection or shoulder small, much slighter than in trigono-
stoma. Just caudad of the tentaculiferous fold are numerous minute dark
eye-spots. Tentacles crowded, numerous and long; slender; each with the
usual longitudinal groove. Segmental papillae present only on segments
III, IV, and V. Of the thoracic uncinigerous tori the first six bear the uncini
in single simple series, the others in a double, interlocking series. Seventeen
setigerous somites present. Ventral thoracic plates very wide, trapeziform,
the lateral ends being angularly pointed; the anterior ones rather more than
twice as wide as the length of the adjacent row of uncini; caudad the anterior
margin becomes more decidedly convex; on II, III, and IV they are not set
off from the lateral region; these plates roughened; caudad of XVI, plates
reduced and longitudinally divided. First branchiae decidedly longer than

266 bulletin: museum of comparative zoology.

the others, the trunks stouter and longer, with two main limbs and a dense
brush of terminal branches. Total nimiber of segments in type near eighty-
five.

Length, 110 mm.; width of thorax, 11 mm.
Locality. — Calif.: Crescent City (A. Agassiz).
Type.— M. C. Z. 2,135. Paratypes.— M. C. Z. 449.

31. SciONiDEs DUX, sp. nov.

Plate 3, fig. 9.

As compared with the genotype {Terebella reticulata Ehlers) this is a very
much larger species easily differentiated as well by various other characters,
such as the structure of uncini which, while agreeing closely in more general
features, differ in numerous details. These have the basal shoulder more
remote from the crest, thus leaving the neck-region longer; the sinus wider
at bottom, the beak less divergent from principal axis. Differing from those
of reticulata decidedly in having the denticles of the crest in five series instead
of in four with the teeth of these series more numerous, the larger teeth of the
lowermost row, e. g., numbering six instead of only two. There are seven-
teen setigerous segments, bearing setae of simple limbate type, with no trace
of distal denticulations. The uncini begin on the second of these (V); they
are in single series on the first six pairs of tori and in a double interlocking
series on the remaining pairs. There is a non-setigerous tubercle below each
of the second branchiae (somite III), this well developed but much more
slender than the setigerous one. Tentacular filaments numerous, closely
crowded in a series across anterior edge of fold, flat and ribbon formed, long.
No eye-spots observable in types. Peristomium with labial edge below wide
and straight, a second edge farther caudad with elongate pit or depression
between the two. Branchiae three pairs on II, III, and IV. These ramose,
with thick trunks and principal limbs, and numerous fine terminal branches or
filaments densely grouped. The third branchiae are inserted obviously nearer
together than the second and first. Not differing in length but the second
branchiae with more abundant branches. First twelve ventral plates broad,
the first ones oblong, the others becoming trapeziform, the subsequent ones
very narrow. Plate of segment II longer than that of III and both much
longer than those of IV and V, the others increasing in length and decreasing
in width in going caudad. Number of segments in type seventy-five.

Length, 125 mm. ; width across thorax, 10 mm.
Locality. — Calif.: Crescent City (A. Agassiz).
Type.— M. C. Z. 2,034. Paratypes.— M. C. Z. 228.

chamberlin: pacific coast polychaeta. 267

Six specimens from Mendocino (M. C. Z. 75) are also referred to this
species. They agree closely in most features though the branchiae
are shorter and the uncini are apparently slightly more slender.

32. Thelepus crispus Johnson.

Proc. Boston soc. nat. hist., 1901, 29, p. 428, pi. 17, fig. 175-178b.

Several small specimens of this species were collected by Mr. Agassiz
at Crescent City, others at San Francisco, and many at Mendocino.

Sabellidae.

33. EuDiSTYLiA POLYMORPHA (Johnson).

Bispira polymorpha Johnson, Proc. Boston soc. nat. hist., 1901, 29, p. 429, pi.
17, fig. 179-183; pi. 18, fig. 184, 185.

Five specimens of this form were taken by Mr. Agassiz in the Gulf
of Georgia, Washington, and preserved free from their tubes. Three
of these are exceptionally large, one having a width across thorax of
nearly 20 mm. In addition to these specimens (M. C. Z. 485) there
are several specimens in situ in their tough cartilaginous tubes (M. C. Z.
486).

34. DiSTYLIA MONTEREA, Sp. nOV.

In comparing Mr. Agassiz's Gulf of Georgia specimens of E. polymorpha with
Johnson's types of that species it was noted that a specimen from Pacific
Grove labeled by Johnson as one of the paratypes was not conspecific or even
congeneric with the others. In size, form, coloration, and general appearance
it is remarkably similar to polymorpha, but that it is really generically distinct
is at once shown by the fact that the inferior setae of the collar-fascicle are
lanceolate instead of spatulate in form. These setae are also much fewer in
number than in polymorpha. The inferior thoracic setae of the other seg-
ments, however, are of the usual spatulate form. Another readily noted differ-
ence, which at the same time separates this species from Distylia rugosa
(Moore), is that there are only seven setigerous thoracic somites instead of
eight. Whereas in rugosa the eyes are very numerous, approximating one
hundred on each radiole, in the present species they are very sparse, most
radioles lacking them entirely; when present they are usually two on a radiole

268 bulletin: museum of comparative zoology.

and widely separated, occasionally three. The collar is prominent with the
general dorsal opening wider than in rugosa, equalling fully three fourths the
body width at that level. The dorsal lobes very low in proportion to width.
The ventral lobes in the type are widely separated by a V-shaped opening,
not contiguous or overlapping as in polymarpha. The general coloration is
that of polymorpha, the branchiae having a similar deep wine-color with lighter
transverse bandings distally. Number of segments in type, near two hundred
and twelve.

Length without branchiae, 140 mm. ; greatest width, 12 mm.
Locality. — CaHf. : Pacific Grove (H. P. Johnson).
Tijpe.~M. C. Z. 1,941.

35. Pseudopotamilla paurops ChamberUn.

Pomona college journ. zool. ent., 1919, 11, p. 21.

A single specimen, now unfortunately dry, taken at Mendocino>
(A. Agassiz) agrees with this species so far as the characters are evi-
dent. Previously known from Laguna Beach, Calif.

)
36. Pseudopotamilla brevibranchiata Moore.

Proc. Acad. nat. sci. Philad., 1905, p. 555, pi. 37, fig. 1-7.

Several dry specimens seem to conform to this species, though the
branchiae are rather longer than indicated in the original description.
The branchiae agree in structure and number and have the eyes
similar in prominence, number, and distribution over the middle
half of the radioles. The uncini agree Aery well, though in those of the
sixth segment the neck appears to be a little more slender; the beak
and crest together have the characteristic straight anterior edge with
the tip of beak slightly bent forward. The specimens were taken at
Mendocino (A. Agassiz).

37. Pseudopotamilla panamica, sp. nov.

Plate 3, fig. 8.

In the type of this form there is a total of sixty-seven somites of which nine
(eight setigerous) are thoracic. The branchiae are transversely banded with
dark, the banding dusky and not sharply limited. The branchial membrane

^ chamberlin: pacific coast polychaeta. 269

crossed longitudinally by purplish lines, one being opposite the interval
between each two radioles. On the thorax a broad purplish brown longitudi-
nal band mesad of and more or less embracing the notopodia on each side, a
yellow median dorsal stripe between the two dark bands, the latter fading
out caudad. Ventral plates pale orange, the anterior thoracic ones darkened
at each lateral end, a dark band extending up each side of the first segment
and a dark line in front and one behind each of the anterior tori. Ventral
lobes of collar not extending much forward, broadly overlapping in the middle
line. The dorsolateral incision on each side deep and narrow; each dorsal
lobe extending forward beyond the lateral part, oblong, with ectoanterior
corner and anterior end convexly rounded, the anteromesal corner subrec-
tangular, the mesal edge a little concave, the ectal gently convex. Free dorsal
margins of branchial basal membrane not at all incised or lobed, well separated.
No eyes present. Radioles twenty-two on each side, in a simple series. All
thoracic plates and the first abdominal one entire, the other abdominal plates
longitudinally bisected. Thoracic uncini characterized by an unusually long
and erect neck, the lower protruding lobe small, the beak straight and widely
diverging. Inferior thoracic setae distally spatulate with a slender acute tip
usually curving more or less at an angle with the blade.

Length, 35 mm.; greatest width, 3.5 mm.
Locality. — Panama.
Type.— M. C. Z. 72.

The tube proper is tough and parchment-Hke with outer surface
densely coated with shell fragments and sand.

Serpulidae.

38. Serpula vermicularis Linne.
Syst. nat., ed. 12, 1767, 1, p. 1267.

Serpula vasifera Haswell, Proc. Linn. soc. N. S. W., 1885, 9, p. 608, pi. 31,

fig. 1, pi. 32, fig. 6-8.
Serpula columbiana Johnson, Proc. Boston soc. nat. hist., 1901, 29, p. 432,

pi. 19, fig. 199-204.
Serpula narconensis CoUin, Semon's forsch. Austr. Malayen archipel., 1902,

p. 100.
Serpula granulosa Willey, Ceylon pearl oyster fisheries report. SuppL, 1905,

p. 316, pi. 7, fig. 186, 186a.

Several specimens of this widespread form, of which the synonomy
pertaining to the Pacific region only is given, with their bright white
tubes were taken by Mr. Agassiz at Mendocino, and in the Gulf of
Georgia, Wash.

270 bulletin: museum of comparative zoology.

39. Serpula nannoides, sp. nov.

Plate 2, fig. 8.

The present form, as represented by the several type-specimens, is greatly
exceeded in size by S. vermicularis and S. splendens. From both these species
readily distinguishable in having the collar-fascia relatively much longer and
more prominent, greatly exceeding the following fasciae instead of being
smaller than them. Collar-setae stouter and darker than the others and
differing in structure as usual; bayonet forms with spurs stouter than in
vermicularis, rounded. Thoracic membrane at lower end below on each side
with two caudally directed, triangular lobes or flaps of which the lower is
usually much larger, the upper one sometimes scarcely obvious; on ventral
side the collar-membrane protrudes forward at middle where it is simply
rounded, not at all incised. Operculum funnel-shaped, thin and less rigid
than in vermicularis, more or less readily collapsible. Number of crenations
along edge of operculum about ninety. Secondary operculum not observed
in the types.

Greatest width, near 2.5 mm.

Locality. — Calif.: Crescent City (A. Agassiz).

Type — M. C. Z. 511. Paratypes.— M. C. Z. 2,131.

40. Crucigera hespera, sp. nov,

Plate 2, fig. 9.

The type of this species is notably smaller than that of C. zygophera (John-
son) with which it has been compared. It is obviously different in the form
of the operculum. The operculum proper is similarly regular but is decidedly
narrower, not truly campanulate, the radii not flaring out distally but with
their acute apices in a distal median position on each. Radii twenty-five in
number. As in zygophera there is on one side but a single rounded, ectally
flattened or concavely depressed lobe beneath which the distally abruptly
constricted stalk is attached. On the other side are two' more prominent
lobes; these are more widely divergent than in zygophera and instead of being
straight are distally geniculate, the short apex extending up at right angles
to the axis of the basal part and well rounded. Secondary operculum more
slender, distally less bluntly rounded than in zygophera.

Width across thorax, 2.5 mm. Width across operculum, 1.8 mm.
Locality.— C alii.: Mendocino (A. Agassiz).
Type.— M. C. Z. 164.

EXPLANATION OF THE PLATES.

PLATE 1.

Chamberlin. — Pacific Coast Polychaeta.

PLATE 1.

Hesperonoe senilis Chamberlin.

Fig. 1. Notopodial seta of first type. X 152.

Fig. 2. Notopodial seta of second type. X 152.

Fig. 3. Neuropodial seta of first type. X 152.

Fig. 4. Neuropodial seta of second type. X 152.

Nereis mendocinana Chamberlin.
Fig. 5. Dorsal view of prostomium. X 47.

Leodice valens Chamberlin.

Fig. 6. Composite seta. X 300.
Fig. 7. Mandibles. X 14.
Fig. 8. Maxillae I. X 14.

BULL. MUS. COMP. ZOOL.

Chamberlin. Polychaeta. Plate 1

,: N

PLATE 2.

Chamberlin. — Pacific Coast Polychaeta.

PLATE 2.

PisTA BREViBRANCHiA Chamberlin.

Fig. 1. Thoracic uncinus of segment V. X 616.

Fig. 2. Thoracic uncinus of segment XI. X 616.

Fig. 3. Thoracic uncinus of segment XX. X 616.

Fig. 4. Abdominal uncinus.

Sabellaria nanella Chamberlin.

Fig. 5. Opercular paleolus of outer series, caudoectal view. X 80.
Fig. 6. Opercular paleolus of inner series, lateral view. X 80.
Fig. 7. Parathoracic notopodial seta. X 244.

Serpula nannoides Chamberlin.
Fig. 8. Operculum. X 20.

Crucigera hespera Chamberlin.
Fig. 9. Operculum. X 20.

Nainereis nannobranchia Chamberlin.
Fig. 10. Thoracic neuropodial seta of posterior series. X 180.

BULL. MUS. COMP. ZOOL.

Chamberlin. Polychaeta. Plate 2

PLATE 3.

Chamberlin. — Pacific Coast Polychaeta.

PLATE 3.

Nainereis nannobranchia Chamberlin.
Fig. 1. Thoracic neuropodial seta of anterior series. X 180.

Idanthrysus ornamentatus Chamberlin.

Fig. 2. Opercular paleolus of inner series. X 38.

Fig. 3. Opercular paleolus of outer series. X 38.

Fig. 4. Thoracic paleolus. X 38.

Fig. 5. Abdominal uncinus. X 616.

EupoLYMNiA crescentis Chamberlin.

Fig. 6. Thoracic uncinus. X 616.
Fig. 7. Abdominal uncinus. X 616.

PsEUDOPOTAMiLLA PANAMiCA Chamberlin.
Fig. 8. Thoracic uncinus. X 616.

SciONiDES DUX Chamberlin.
Fig. 9. Thoracic uncinus. X 616.

BULL. MUS.COMP. ZOOL.

Chamberlin. Polychaeta. Plate 3

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LXIII. No. 7.

THE ANTS OF THE BRITISH SOLOMON ISLANDS.

By William M. Mann.

With Two Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MQSEUM.

December, 1919.

No. 7.— The Ants of the British Solomon Islands.
By William M. Mann.

Introduction and Itinerary.

During the years 1915-1916 the writer was the holder of a Sheldon
Travelling Fellowship of Harvard University and funds from this
were used in zoological exploration of certain Soiith Sea Islands.
Later the Museum of Comparative Zoology contributed a sum of
money, which, with funds from other sources, enabled me to continue
the work and to visit the British Solomon Islands, where I remained
from 19 May to 24 November, 1916.

Upon my arrival and throughout my entire stay I was the recipient
of so many courtesies and so much aid from so many people that to
enumerate them would be practically to make a census of the white
population of the islands.

Some time was spent on Florida, where Dr. O'Sullivan, the port
doctor, welcomed me to his home at Tulagi and stored my specimens
and supplies while I was in the out-islands.

Malaita was next visited. Mr. W. R. Bell, with whom I stayed at
Auki, the government station of which he is in charge, made many
excursions with me into the little known interior and some months
afterward, on my return to Auki, we crossed the island, camping one
night in the mountains, where, at an altitude of about 2,300 feet, we
encountered a number of species not found elsewhere, among them
Crematogaster {Rhachiocrema, subg. nov.) wheeleri, sp. nov. and the
curious Polyrhachis {Dolichorhachis, subg. nov.) malamsis, sp. nov.
The native constabulary who accompanied us, necessary companions
in the forests of Malaita, proved to be excellent collectors and to them
both on Malaita and San Cristoval, I am indebted for many valuable
specimens.

Mr. Symmington, Manager for the Lever's Pacific Plantations Com-
pany, permitted me to travel on the Kobiloko, a small steamer that plies
between certain of the islands, and this made possible a five days stop
at the remote Santa Cruz Group. Here Mr. Jack Mathews arranged
with the natives so that I was able to collect not only unmolested by,
but with considerable help from them. On the return of the Kobiloko I
landed at Ugi, the type-locality for many of the Solomon Island species,
and remained there for several weeks at Pawa as guests of Messrs.

274 bulletin: museum of comparative zoology.

Hall and Dickerson; also visiting the nearby island of Malapaina,
one of the Three Sisters Group, and staying there with Mr. Ireland,
a young Australian engaged in clearing the forests for a new planta-
tion. At Pamua, on the mainland of San Cristoval, I lived with the
Rev. Mr. Nind, at the school maintained there b^^ the Melanesian
Mission. Their steamer, the Southern Cross called and through the
•courtesy of the Rev. Mr. Wood, Bishop of Melanesia, I received a
passage to Wainoni.Bay. Here the two French priest-missionaries,
Fathers Moreau and Babbiau, cared for me. For their great kindness,
especially when stricken with fever, I am deeply grateful.

Mr. Harry Jacobsen, a planter and trader, took me from Wainoni
Ray to Star Harbor, his station at the extreme eastern end of the
island, and made numerous trips with me along the coast and to the
neighboring island of Santa Anna, and afterw^ards up the coast to
Keri Keri, where I met Captain F. M. Campbell, Director of
native constabulary, and returned to Tulagi with him. Then Mr.
Abbott, Government Labor Agent, invited me to accompany him on
his launch to the Russell Islands. In New Georgia I lived with Mr.
Norman Wheatley, Nestor of South Sea traders, travelled about the
beautiful, though gruesomely historic, Rubiana Lagoon with him
and also stopped at Rendova for a few days with Mr. Palmer.

Toward the end of my stay, I collected for three weeks at Fulakora
on the western end of Ysabel, living at the plantation of Mr. Charles
Bignell, a gentleman who spent much time collecting with me and
after my departure sent me a considerable number of interesting
species.

To all of these gentlemen and to others, I am greatly indebted for
making my stay in the Solomons not only safe and successful, but
immensely enjoyable. Without their aid I could have done compara-
tively little collecting.

The greater part of the country is heavily wooded and, as might be
■expected, a large proportion of the species are arboreal. The char-
acter of the fauna changes markedly when the forest is cleared and in
the plantations one is impressed by the scarcity of endemic forms,
which have been supplanted by introduced species. Every log or
board on the ground shelters formicaries of tropicopolitan ants, chiefly
Plagiolcpis longipes and Prenolcpis longicomis. A few Oriental species,
as Oecophylla smaracjdina and Technomyrmex albipcs, are apparently
able to hold their own and Camponotiis reiicuUdus bedoti is not uncom-
mon in cleared land, but most of the native species are doomed as

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 275

soon as the forest goes. The above tramp species do not usually
wander far into the forest, though Plagiolcpis longipcs is sometimes
seen there under conditions noted hereafter.

It is interesting that Solenopsis gcminaia rnfa has not yet established
itself in the Solomons. Neither does it occur in Fiji, though it is
common in both New Guinea and in Tahiti.

One faunal peculiarity is the prevalence of white-tipped antennae
among the ants, species of three genera Wheeleripone, Crcmatogaster
{Rhachiocrema) whederi and Polyrhachis {P. tih/ssrs, P. o.s'flf) having
this unusual, among ants, coloration.

The ant fauna of the Solomons has been comparatively untouched.
Mr. W. W. Froggatt, who visited the islands for the purpose of study-
ing the insects of the coco-nut palm, collected a few species at Tulagi,
on Guadalcanar, and in the Russell Group. These, recorded by Forel
in a paper on Australian ants (Rev. Suisse zool., 1910, 18, 1-94), are
the only species listed from the Solomons. In the present paper are
noted 144 forms, representing forty-three genera.

Ants abound in the forests. Along the paths one sees species of
Polyrhachis of the rcluccns or daemeli groups on nearly e\ery tree and
shrub. In the eastern islands the leaves of single trees sometimes hold
a half dozen nests of such forms as Polyrhachis osar and P. mncronata
mahu'tms. A rotten log will almost certainly contain colonies of T 'ollcn-
hovia pedcstris, Phopalofhrix malua, Myopopone castanca, or Rhytidopo-
nerafroggatti. On the beaches if one leans against a tree he will proba-
bly be covered by swarms of Iridomyrmcx myrmccodiac which run about
and attempt to bite in an annoying manner very similar to species of
Azteca in the American tropics; among trees that have recently been
felled, Occ.ophylla smaragdina nitida makes collecting a misery by its
numbers and aggressiveness.

The last few days of collecting pelded a number of genera and spe-
cies not taken previously and I am convinced that the present list
contains only a small part of the species that occur in this island group.
The interior especially will be producti^ e of many additional forais of
great interest, for with the exception of the few that I list from the
interior of Malaita, the ant fauna of the highlands is absolutely
unknown.

As far as the zoogeographical position of the Solomons is concerned,
the distribution of the ants merely corroborates the opinion of Wallace
who, (The Malay Archipelago, 18G9, 2, p. 435) on the basis of the
presence of crimson lories and of cockatoos allied to those of New
Guinea, assumes that the fauna is a continuation of the Papuan.

276 bulletin: museum of comparative zoology.

Of the forty-three genera recorded in the present paper, none are
pecuHar (Wheeleripone occurs also in Fiji). Typical Australian gen-
era are represented only by Rhytidoponera, Podomyrma, Turneria,
and Opisthopsis, each with a single species.

The remaining genera are all widely distributed in Indo-Malaya.
Of these the following terminate their eastward distri})ution in the
Solomons : — Myopopone, Platythyrea, Cryptopone, Ectomomyrmex,
Crematogaster, Myrmecina, and Acropyga.

Other genera, as Euponera, Anochetus, Vollenhovia, Tetramorium,
Pristomyrmex, Triglyphothrix, Oligomyrmex, and Polyrhachis are well
developed in the Papuan region, but extend eastward of the Solo-
mons in onl}' a few species, which are usually widely distributed.

Two species of wide distribution, Lcptocjcnys dimhniia and Occo-
■phyUa smaragdina, terminate their natural distribution in the Solo-
mons, for though the latter species occurs in Samoa, it is of recent
introduction.

In general the Solomon Island ant fauna is Papuan, without pecu-
liar forms, and lacking a number of typical Papuan genera. The
Santa Cruz ant fauna is similar to that of the Solomons proper and
may be considered as the eastern limit of the Papuan subregion.

The nearby island of Vanikoro has not been explored entomologically
but forests of Araucarians are known to occur there. This is then the
most northwestern locality for the Araucarinae and the island con-
taining it probably belongs to the Melanesian subregion, so a line
between the two islands separates the two subregions.^

In the preparation of this paper, I have been greatly assisted by
Prof. W. M. Wheeler of Harvard University, who has generously
permitted me to use his collection for comparison and has aided me in
certain determinations and in other ways. Prof. Carlo Emery, of
Bologna has kindly compared a number of species with types in his
collection.

Most of the figures of new species were made by Dr. R. McEwen.

1 The Melanesian subregion, (including Vanikoro, the New Hebrides, New Caledonia and
Fiji), like the Chilean and the Malagasy, might be described as a biological conservatory, where
types once tropicopolitan have been isolated and preserved, free from invasion. It is distinct
from Pajmasla though the presence of numbers of these relicts, and in negative characters,
lacking the majority of the plant and animal groups characteristic of the latter subregion. The
absence of any group which had its origin in this region is an important negative character.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 277

Formicidae: Ponerinae.

la. Cerapachys (Syscia) pusilla Emery, subsp. pawa, subsp. nov.

A single worker found beneath a stone agrees with Emery's descrip-
tion of ixipuana from New Guinea, except that it is smaller and less
opaque and has the postpetiole distinctly longer than broad for which
reason I refer it to the closely related pusilla Emery, also from New
Guinea, which differs from papuana in the same characters.

The subspecies pawa differs from pusilla only in sculpture. The
head is irregularly and coarsely punctate and rugose. The thorax is
shining and more regularly punctate and not rugose.

The legs are shining.

Ugi : Pawa.

2. Cerapachys (Cerapachys) terricola, sp. nov.

Worker. Length 3.5 mm.

Head a third longer than broad, as broad in front as behind, with moderately
rounded sides, nearly rectangular occipital corners and feebly concave occipital
border. Mandibles short and stout, blades minutely denticulate. Frontal
lobes short, narrowly separated. Anterior border of clypeus almost straight.
Antennal scopes thick, club-shaped, extending more than half the distance to

Fig. 1.- — Cerapachys (Cerapachys) terricola Mann. Worker.

occipital corners; first flagellar joint as long as broad; joints 2-11 transverse;
apical joint as long as the six preceding joints and at the middle one and a half
times as thick as the penultimate. Eyes large, convex, located at sides of
head a little anterior to the middle. Thorax two and a half times longer than
broad; sides straight, anterior border evenly rounded and narrowly margined;

278 bulletin: museum of compakative zoology.

in profile convex above. Epinotal declivity flat, margined above and at sides.
Petiole subrectangular, a little broader than long, margined in front, sides
feebly rounded, anterior surface flat; in profile, about as long as high, rounded
above, ventral tooth large and triangular. First gastric segment similar in
shape to petiole, about a third broader ; remainder of gaster less than twice as
long as broad.

Mandibles sparsely punctate and shining. Head, thorax, and gaster with
scattered foveolate punctures, which are more abundant and largest on the
head, thoracic pleurae, and petiole, becoming contiguous on the metapleurae,
more sparse on the pronotum and gaster, and lacking on epinotal declivity.
The interspaces smooth and shining. Legs and antennae finely punctate and
sublucid.

Head and body with fine, long, erect hairs; legs and antennae with shorter,
stiffer, and semierect hairs. Blades of mandibles with brush of short pile.

Color piceous; mandibles, legs, and antennae ferruginous, tibiae partly
fuscous.

Male. Length 3 mm.

Head, excluding eyes, longer than broad, narrowly rounded behind; cheeks
short. Mandibles very similar to those of worker. Clypeus as in worker.
Antennae short, scapes club-shaped, not extending to occipital corners; funic-
ular joints longer than broad, gradually increasing in thickness toward apex;
apical joint a little longer than the two preceding together and but slightly
thicker than the penultimate. Eyes and ocelli large and convex. Promeso-
thorax in profile strongly convex, without Mayrian furrows. Mesometanotal
suture straight, distinctly impressed, scutellum convex. Epinotum convex,
without distinct base and declivity. Petiole and first gastric segment similar
to those of worker. Gaster thick. Genitalia prominent.

Punctation similar to that of worker, but more feebly sliining. Pilosity as
in worker.

Wings deeply infuscated and densely covered with short suberect hairs.
Stigma dark fuscous.

Piceous; mandibles and antennae fuscous; tibiae dark testaceous.

Three Sisters: Malapaina (Type-locaHty). Ysabel: Fulakora.

Described from several workers and males taken from a colony
situated in wet earth beneath a stone and from two workers found
running on the ground. Type. — M. C. Z. 9,151.

The distinctly, though finely, denticulate mandibular blades and
the more slender antennae distinguish terricola from inconspicua Emery.
In the latter species the second funicular joint is very small; the
funiculus robust, with the club barely longer than the four preceding
joints. In terricola the club is as long as the six preceding joints.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 279

2a. Cerapachys (Cerapachys) terricola, subsp. tulagi, subsp.nov.

Worker. Length 3.5-4 mm.

Differing from typical terricola in its larger size and in the sculpture of the
metathoracic pleurae and sides of petiole, where the punctures are more
shallow and sparse and not confluent. In terricola the punctation is dense,
the sides of the petiole being almost rugose in appearance.

Male. Length 3.8 mm.

Similar to male of typical form but conspicuously larger in size.

Florida: Tulagi.

One small colony found beneath a stone.

3. Cerapachys (Cerapachys) inconspicua Emery.

Term, fuzet., 1932, 25, p. 152, ^ .

Ysabel: Fulakora.

A solitary worker which agrees with Emery's description was found
running on the ground.

Stigmatomma subgen. Fulakora, subgen. nov.

This subgenus is proposed for those species of Stigmatomma that
have the frontal lobes approximate, instead of widely separated.

Type. — S. (Fulakora) celata, sp. nov.

S. arviigerwn Mayr, chilense Mayr, saundcrsi Forel, and minuta
Forel should be included in Fulakora.

4. Stigmatomma (Fulakora) celata, sp. nov.

Worker. Length 2.9 mm.

Head a little longer than broad; occiput shallowly concave; sides nearly
straight in front, broadest at clypeus, behind somewhat rounding into moder-
ately narrow occipital corners; a feeble suture extending from occiput to

280

bulletin: museum of comparative zoology.

frontal lobes; clypeus in front, armed with six stout denticles. Mandibles a
little less than two thirds as long as head, the thickened basal portion with five
teeth, the apical three bifid ; the slender apical portion with one small denticle
and two minute ones near apex. Frontal lobes convergent, the carinae short,
feeble, and diverging behind. Scape short, extending less than two thirds the

distance to the occipital corners ; all
the funicular joints except the first
and apical distinctly transverse;
apical joint nearly twice as long as
broad. Eye distinct, but minute,
located at sides of head posterior to
the middle. Thoracic dorsum and
epinotum flat. Prothorax a little
longer than broad, evenly rounded
in front; sides parallel; disc at
middle with a very feeble longitu-
dinal carina. Promesonotal suture
strongly impressed. Mesonotum
nearly twice as broad as long, sides
evenly rounded. Epinotum one
and one third times longer than
broad, broadest behind, sides con-
vex at anterior half, then straight;
declivity transverse, strongly mar-
gined at sides, disc shallowly con-
cave. Petiole narrower than first
gastric segment; from above a little
broader than long, rounded, in frorrt, with nearly straight sides; from the side
slightly longer than deep; ventral process small, rounded in profile. Gaster
slender. Legs stout.

Sublucid throughout. Mandibles with short coarse rugae and stiff erect
hairs. Scapes, head, thorax, and petiole densely granulose, the head more so
than the rest, with short, semierect pilosity. Petiole, gaster, metathoracic
pleurae, and legs punctate, similar to, but more shallow than, the rest of the
body; the pilosity of the gaster becoming dense toward apex.

Color ruf o testaceous ; the head a little darker; pilosity yellowish white.

Fig. 2. — Stigmatorama (Fulakora) celaf a
Mann. Worker. Front view of head.

Described from workers taken with larvae and pupae from small
colonies nesting in humid forests beneath stones, on Ysabel (Fulakora
Type-locality), and Malaita (Auki) and from Tulagi, Florida, and
Wai-ai, San Cristoval. The specimens from Tulagi and Wai-ai are
darker than the others. Type. — M. C. Z. 9,152.

Pupae entirely naked.

The larva is covered with fine, white, erect hairs and is very similar to

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 281

that of S. paUipes, described and figured by Wheeler (Biol, bull.,
1900, 11, p. 61, fig. 8) but rather more slender.

5. Amblyopone levidens Emery.

Ann. Mus. civ. stor. nat. Geneva, 1887, 25, p 448, 9-

Malaita: Interior near Fourafi.

A single female agrees with Emery's description.

6. Myopopone castanea (F. Smith).

Amblyopone castaneus F. Smith, Journ. proc. Linn. soc. London. Zool.,
1861, 5, p. 105, pi. 1, fig. 6, y .

Santa Cruz: Graciosa Bay. San Ci'istoval: Wai-ai, Star Harbor.
Florida: Tulagi. Ysabel: Fulakora. New Geoi-gia: Lambeti.

My specimens belong to the typical form,, which is widely distrib-
uted in the Papuasian and Oriental regions and occurs also in iVustralia.

It nests in rotten or semirot'ten wood, always where there is plenty
of moisture. The workers are very timid and highly sensitive to and
repelled by light.

7. Platythyrea melancholica (F. Smith).

Pachycondijla inelancholica F. Smith, Journ. proc. Linn. soc. London. Zool.,
1864, 8, p. 71, y .

Santa Cruz : Graciosa Bay.
One worker.

8. Rhopalopone malaensis, sp. nov.

Worker. Length 2 mm.

Head about a third longer than broad; sides slightly convex; occipital
angles narrow, margin feebly concave. Clypeus broadly rounded in front,
the surface convex. Frontal lamellae small, widely separated. Mandibles
rather slender, the blades with very minute teeth. Antennal scapes extending
almost to the occipital corners; funicular joints 4-9 slightly transverse. Eyes
small, but distinct, situated at sides of head a little in front of the middle.

282 bulletin: museum of comparative zoology.

Promesothorax in profile slightly convex ; from above, narrowest behind, broad-
est in front of middle, with rounded sides and anterior border. Promesonotal
suture obsolete. Mesoepinotal suture barely discernible. Base of epinotum
quadrangular, nearly as broad as long, with straight sides, the surface convex ;
declivity flat, with slightly margined sides. Petiole more than twice as broad
as long, narrowest in front, in profile deeper than long, rounded above. First
gastric segment twice as broad as petiole; a little longer than the second
segment. Legs stout.

Head, thorax, petiole, and gaster sublucid, densely pitted with coarse
irregular foveae which are narrowly separated, so that the body has a reticu-
late appearance; clypeus longitudinally striate. Epinotum strigose with
fewer foveae; declivity smooth and shining. Meso- and metapleurae coarsely,
irregularly strigose. Mandibles coarsely punctate, shining. Antennae and
legs sublucid, finely punctate.

Antennae pubescent. Head and body with fine erect pile. Color dark
fuscous. Antennae, mandibles, and legs ferruginous. Pilosity white.

Malaita: x\uki.

Described from workers taken from beneath the bark of a dead tree.
Type.—M. C. Z. 9,153.

This form approaches R. epinotalis Emery from New Guinea, but
in that species the petiole is not punctate and the third abdominal seg-
ment is subtly punctate. In malarusis, the petiole is coarsely punctate,
and the third abdominal segment is as coarsely so as the second. It
differs also in having denticles, though very small ones, on the mandibu-
lar blades. Possibly it should be considered merely a subspecies of
ciyinotalis.

The larva is slender, not tuberculate, and uniformly covered with
long white hairs, which become fine and flexuous anteriorly, similar
to the larvae of Stigmatomma.

The pupae are light fuscous in color.

Wheeleripone, gen. nov.

Tyjye. — W. albiclava, sp. nov.
Worker. Allied to Stictoponera Mayr.

Head moderately elongate. Mandibles elongate, triangular, with small
teeth on inner border. Frontal lobes small, feebly prominent and widely
separated. Clypeus strongly depressed anteriorly and broadly rounded,
separated from front by a feeble suture. Eyes prominent and convex, though

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

283

small. Antennae slender, 12-jointed; the scape somewhat flattened at base;
joints 8-11 forming a very poorly differentiated club. Thorax elongate; un-
armed. Prothorax angulate ventrally at sides but without tooth. Pro-
mesonotal and mesoepinotal sutures strongly impressed. Petiole from above,
subglobose; from the side, rounded above, the anterior surface with a flattened
space, anteroventral surface with flattened triangular spine. First gastric
segment a little broader, but shorter than the second, armed anteroventrally.
The remaining segments short, somewhat deflected, but not as strongly as in
Stictoponera. Legs slender; anterior tibia with strongly pectinate spine;
middle and posterior tibiae each with a single long, simple spine. Tarsal
claws large, with a distinct pointed tooth basally on the inner surface.

9. Wheeleripone albiclava, sp. nov.

Worker. Length .6.5 mm.

Head a third longer than broad, broadest at occiput; sides slightly convex;
occipital border shallowly and narrowly excavated. Frontal carinae very
short, their lobes small, flat, as far apart as their distance to sides of head.
Basal portion of clypeus slightly convex, longer than broad; anterior portion
flat, the border evenly rounded. Mandibles large, elongate-triangular, the
blades with 10-12 minute, widely separated teeth. Antennae slender, scapes

Fig. 3. — Wheeleripone albiclava Mann. Worker.

somewhat flattened basally, surpassing occipital angles by a distance of less
than their width at apex; first flagellar joint slightly longer than the second
and twice as long as broad, second joint a little longer than the third; joints
3-7 subequal, subglobose in shape; joints 8-10 each a little larger than the
preceding, 9-10 longer than broad; terminal joint slender, a little longer than

284 bulletin: museum of comparative zoology.

the two preceding joints together. Eyes small, but distinct, convex; located
at middle of sides of head. Prothorax slightly convex, broadest in front,
humeri narrowly rounded; anterior border rounded, sides nearly straight;
ventral border angulate, but not armed. Promesonotal suture strongly im-
pressed. Mesonotum more than twice as broad as long; the surface feebly
convex. Base of epinotum slightly convex, little longer than broad ; in profile,
broadly rounding into the declivity; declivity slightly transverse, the surface
convex. Petiole from above subglobose, a little longer than broad; in profile
about as deep as long; anterior surface flat and margined at sides; antero-
ventral tooth flat, triangular. First gastric segment at base a little broader
than the petiole; broader, but shorter than the second segment; the constric-
tion between the two strong. Sting short and thick for two thirds the distance
from base, then acuminate, the sides of the thickened basal portion at tip with
a fine hair. Legs slender; tarsal claws with a strong tooth.

Shining; occiput, sides of head, and cheeks furrowed, the furrows containing
strong, umbulate foveae, sometimes approximate, but not contiguous; middle
of front and vertex with seven strong irregular carinae, the inner of which
extend to the occipital border and the outer only half the distance; basal
portion of clypeus bicarinate, the carinae widely separated, anterior portion
densely punjtulate. Mandibles feebly shining and shallowly punctate
Antennae sublucid, shallowly punctate. Prothorax with a few shallow
punctures on front half of dorsum, elsewhere smooth. Mesothorax with
strong carinae, six on the disc longitudinal, four at sides becoming diagonal.
Epinotum punctuate at upper part of sides. Metapleurae with five strong,
short carinae situated apically. Petiole and first two gastric segments sparsely
and shallowly foveolate. Legs sparsely punctate.

Head, body, and antennal scapes with very sparse, scattered erect hairs.
Antennal funiculus pubescent.

Ferruginous throughout; funicular joints 1-7 a little lighter; joints 8-11
white. Pilosity black.

Ysabel: Fulakora.

Described from four workers found on the ground in the forest.

This very striking ant approaches in general form some of the species
of Stictoponera but it differs in having very strong thoracic sutures
and in not having angulate occipital corners and the gaster is
much less deflected at tip. The shining integument, with its curious
sculpture and the remarkable coloration of the antennae, ferruginous,
with a white club in strong contrast, is quite distinctive. The club
is perhaps not more pronounced than in Stictoponera, but is strongly
accentuated by the color.

It is evidently rare for I searched diligently in the same vicinity
without finding a nest or more workers.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 285

10. Wheeleripone lucida, sp. nov.

Worker. Length 3 mm.

Differing from the preceding species in the much smaller size and in the form
of the petiole, which in profile is distinctly higher than long, convexly declivous
in front, with the flat anterior space very poorly defined.

There are no striae on the mesonotum, which is very smooth and shining.
Otherwise it is a miniature of albidava in form, sculpture, and color, but with
more abundant and proportionately coarser hairs on the head and body.

Malaita: x\iiki.

Described from a solitary worker.

11. Wheeleripone crenaticeps, sp. nov.

Worker. Length 3 mm.

Head one and a half times longer than broad, nearly as broad in front as
behind, with feebly convex sides and narrowly rounded occipital corners;
occipital border very shallowly and narrowly concave at middle. Mandibles
stout, trigonal, the blades with indistinct widely separated denticles. Clj^jeus
convex basally; anterior border subangulate at middle. Frontal carinae

Fig. 4. — Wheeleripone crenaticeps Mann. Worker. Lateral view of thorax and petiole.

strong, parallel for half their length, then slightly diverging and becoming
tortuous. Frontal area deeply impressed. Antenna! scapes surpassing
occipital corners by one fourth of their length; funicular joints 1-2 longer than
broad, the first the longest, joints 3-7 very slightly broader than long; joints
8-11 forming a rather slender club with the terminal joint almost as long as
the other three together. Prothorax broader than long, rounded in front and
sides; in profile convex in front and rather flat behind. Promesonotal suture
very feebly impressed. Mesonotum more than twice as broad as long, flat

286 bulletin: museum of comparative zoology.

above. Epinotum as broad in front aS behind, one and a half times longer
than broad; basal surface flat. Petiole very little broader than long, evenly
rounded in front and at sides; posterior border straight; node in profile,
higher than long, rounded above, declivous behind and nearly so in front; the
anteroventral spine large. First gastric segment broader than long and
noticeably shorter than the second. Legs slender.

Head sublucid, the front and vertex with coarse, sinuous carinae, and the
spaces between those with coarse, foveolate punctures; sides foveolate, the
interspaces smooth and shining. Clypeus bicarinate and finely punctate.
Mandibles sublucid; coarsely but shallowly punctured. Antennae more
finely punctate. Thorax shining, sparsely punctate, the punctures foveate,
but more shallow than those of the head. Petiole and gastric segments
punctate even more sparsely and shallowly. Metapleurae transversely striate
apically. Legs shining, with sparse punctation.

Head and body with fine erect hairs.

Color dark brown; antennae and legs lighter.

Ysabel : Fulakora.

Described from several workers taken from a colony beneath a stone.
Type.— M.C.Z. 9,154:.

Key.

Length 6 mm. Mesonotum strongly striate; petiole in profile longer than
high albiclava Mann.

Length 3 mm. Mesonotum not striate; petiole in profile not longer than

high 1 .

1. Last four funicular joints white; promesonotal suture stronger; pro-
thorax smooth and shining; anterior face of petiolar node shorter than
dorsal surface and broadly rounding into it lucida Mann.

Antennae unicolored; promesonotal suture weaker; prothorax heavily
punctate; anterior face of petiolar node longer than dorsal surface and
narrowly rounding into it crenaticeps Mann .

12a. Rhytidoponera (Rhytidoponera) araneoides Le Guillou,

var. froggatti Forel.

Rev. Suisse zool., 1910, 18, p. 10, t^ .

Male. Length 7-8 mm.

Head, longer than broad, as broad in front as behind, with rounded occiput.
Eyes and ocelli very large and convex; cheeks a fifth as long as eye. Clypeus

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 287

convex, flattened anteriorly, the border evenly rounded. Mandibles stout,
dentate similar to those of worker. Antennal scape extending two thirds the
distance to occipital corners. First flagellar joint one fifth the length of the
second, which is as long as the scape; joints 3-12 gradually shorter than the
preceding; terminal joint slightly longer than penultimate. Thorax robust;

Fig. 5. — Rhytidoponera (Rhytidoponera) araneoides froggatli Forel. Worker.

scutellum prominent, convex, a little broader than long. Declivity of epino-
tum longer than the base and flattened. Petiole more than twice as long as
high, convex above at middle, from above, longer than broad, sides slightly
convex. Genitalia not prominent.

Shining, head and thorax irregularly, rugosely sculptured. Epinotum and
petiole transversely strigose. First segment of gaster very densely and finely
arcuately striate; striae of second segment similar, but more concentric.
Mandibles longitudinally strigose.

Head and body with moderately abundant fine, erect hairs.

Color fuscous throughout; mandibles, antennae, and tarsi paler. Pilosity
black.

Wings (length 5.7 mm.) infuscated.

Ysabel: Fulakora. San Cristoval: Pamua, Wainoni Bay. Ugi:
Pawa. Florida: Tulagi, Maliali. Malaita: Auki, near Fourafi, in
the interior. Rendova.

There appear to be no local varieties of this species which ranges
throughout the group. Probably it is confined to the British Solo-
mons, for it is very common where it occurs and would have been
recorded from further east if found there. I did not find any other
species of this genus, which is well developed in New Guinea.

R. froggaUi lives generally in or beneath hollow logs lying on the
ground, sometimes in colonies numbering several hundred but more
frequently with less. It is a timid species, hustling its dark brown
pupae and the larvae into the nearest shelter and huddling there when

288

bulletin: museum of comparative zoology.

the nest is disturbed. Those individuals not completely hidden remain
motionless. It is fond of plant-juices and workers are commonly
seen on smaller plants.

Excepting the uniformly larger size and the black color, I can detect
no difference that will separate the workers oi froggatti from araneoides
described originally from Bougainville.

13. DiscoTHYREA CLAVicoRNis Emery.
Term, fuzet., 1897, 20, p. 593, pi. 15, fig. 39, 40, ^ .

A small colony containing three females and a dozen workers was
found beneath a log in a swamp near Fulakora, Ysabel. The worker
is more active than those species of Proceratium and Sysphincta that I

Fig. 6. — Discothyrea clavicornis Emery. Wing of female.

have seen. The rich brownish red color, with the microscopic silky
pubescence give the living insect an elegant, shagreened appearance.

The female measures 2 mm. in length and is rather more robust,
but otherwise similar to the worker, except for the usual sexual differ-
ences. The eyes are larger and somewhat convex. The ocelli are
distinct, arranged in an equilateral triangle. The wings (length 2 mm.)
are hyaline and densely covered with fuscous hairs; veins and stigma
are fuscous.

14. EcTOMOMYRMEX EXARATA Emery.
Term, fuzet., 1902, 25, p. 156, ^ .

I am referring to this species three workers. The mandibles are
heavily strigose, with 6-7 teeth on the blades. The third funicular

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

289

joint is longer than broad and the remaining joints are as long as broad.
In the closely related E acuta Emery the third joint is described
as being as long as broad, the remaining apical joints, except the
terminal are transverse and the mandibular blades have only five teeth.
The two species are evidently very closely related and it is probable
that Forel's dalili from the Bismarck Archipelago is only a subspecies
of, if not identical with, rxarafa.

In the specimens before me the first gastric segment is heavily
strigose longitudinally and the second segment is much more delicately
sculptured, as described in the three forms mentioned above.

Ysabel : Fulakora.

14a. EcTOMOMYRMEx EXARATA Emery, subsp. aequalis, subsp. nov.

Worker. Length 6-6.5 mm.

Differing from the t^TDical form in its somewhat smaller size and in having
the striae of the second gastric segment not appreciably finer than that of the
first, though with fewer punctures between.

Fig. 7. — Ectomomyrmex exarata aequalis Mann,
and petiole.

Worker. Lateral view of thorax

Described from several workers taken at Auki, Malaita and one at
Tulagi, Florida. The Auki specimens were together and evidently
belonged to the same colony, but I was unable to locate the nest.
Typc.—M. C. Z. 9,155.

The species of Ectomomyrmex in their timid behaviour are similar
to Bothroponeta.

290 bulletin: museum of comparative zoology.

15. EuPONERA (Mesoponera) papuana Viehmeyer.

Zool. jahr., 1914, 37, p. 608, S .

Worker. Length 6.5-8 mm.

Head one and three fourth times longer than broad, as broad in front as
behind; sides subparallel; occipital angles narrowly rounded, the border
feebly concave. Clypeus very broadly and prominently carina te, the anterior
border angulate at middle. Mandibles long and slender, with eleven stout
teeth. Frontal carinae very short; their lobes little dilated and flat; the
narrow frontal depression extending less than half the distance to occipital
border. Antennae long and slender; scapes surpassing occipital corners by a
distance equal to twice their breadth at apex; first and second funicular joints
subequal, three times longer thfin broad; third joint shorter than second and
proportionally broader, joints 4-10 longer than broad, subequal in length but
increasing in thickness toward apex; terminal joint barely as long as the two
preceding joints together. Eyes small, very flat, situated in front of sides of
head at a distance of three times their diameter from base of mandibles. Pro-
thorax convex, a little broader than long, narrowly rounded at front and sides;

Fig. 8. — Euponera (Mesoponera^ papuana Viehmeyer. Worker. Lateral view of thorax
and petiole.

anteroventral corners broadly rounded. Premesonotal suture strongly, but
less profoundly impressed than the mesoepinotal. Epinotum in profile as
long as prothorax, depressed at middle, with a narrow, transverse impression
slightly posterior to spiracles; angle between base and declivity broadly
rounded and strongly longitudinally impressed at middle, the sides standing
out as rounded margins; declivity flat. Petiole squamiform, twice as high as
long, narrowly rounded above; anteiior surface convex, posterior flat; sub-
marginate above and at sides of posterior border. Gaster long and slender.
Legs very long and slender. Sting slender.

Sublueid. Head, body, mandibles, antennae, and legs finely, reticulately
punctate and micro-scopically pubescent; the punctation and pubescence of

mann: ants of the British solomon islands. 291

the epinotum and petiolar node less conspicuous than on other parts; pronotum
with two long, erect hairs.

Color ferruginous throughout; legs lighter than the rest.

San Cristoval: Wai-ai. Ysabel: Fulakora.

Described from workers taken with larvae from a small colony in a
rotten log.

This is the only f onn of this species described from the Papuan region ;
a female of an undetermined species from New Guinea has been re-
corded by Emery, and other species undoubtedly occur there, as E.
melcmaria Emery with its subspecies is found in the Oriental and
Malayan regions and occurs also in Australia.

Euponera pajmana is sharply distinct from the related species in the
very long, depressed epinotum, with the longitudinal impression at the
apical part of base. It is much larger than rubra F. Smith from Singa-
pore and Java, and more slender than melanaria subsp. australis Forel,
from Australia.

16a. Euponera (Brachyponera) luteipes Mayr, var. crocei-

CORNIS Emery.

Term, fuzet, 1900, 23, p. 315, ^ .

Ysabel: Fulakora.

Several workers were found beneath bark.

17a. Euponera (Trachymesopus) stigma Fabricius, var. c.»uadri-

dentata (F. Smith).

Ponera quadridentata F. Smith, Journ. proc. Linn. soc. London. Zool., 1859,
3, p. 143.

Santa Cruz: GraciosaBay. San Cristoval : Wai-ai, Pamua, Wainoni
Bay. Ugi: Pawa. Malaita: Auki. Florida: Tulagi. New Georgia:
Lambeti. Rubiana Lagoon. Ysabel: Fulakora.

This is as common throughout the Solomons as the typical form is
in the Neotropical region. Both species nest in small colonies in
rotten wood or beneath bark.

292 bulletin: museum of comparative zoology.

18. EupoNERA (Trachymesopus) sheldoni, sp. nov.

Worker. Length 3.75 mm.

Head a little longer than broad, with feebly convex sides and narrowly
rounded occipital corners; occipital border narrowly but rather deeply con-
cave at middle. Cl3T)eus broadly rounded in front; strongly carinate at
middle, the carina produced in front to form a strong, blunt spine. Mandibles
with six stout, triangular teeth. Frontal lobes triangular, flat; the frontal
impression between extending to occipital border. Antennal scapes not
reaching occipital corners; the funicular portion without trace of club; the
joints except first and terminal gradually increasing in size, each only slightly
transverse. Eyes very minute; located in front of sides at about one fifth
the distance from mandibles to occipital corners. Prothorax a little broader
than long; rounded above, in front, and at sides. Mesothorax transversely
oval; in profile almost flat. Promesonotal and mesoepinotal sutures strongly
impressed. Ease of epinotum flat, broadest at middle, as broad in front as
behind; declivity flat, roundly margined at sides. Petiolar node thick; in
profile highest at front , anterior surface concave , apex gradually sloping into
the front surface and broadly rounding into the convex posterior surface:
from above, rounded at front and sides, straight behind, semicircular in shape;
less than twice as broad as long. Basal surface of first gastric segment flat.
Constriction between first and second segments strong.

Head, thorax, and epinotum somewhat shining; petiole and gaster more
strongly shining. Mandibles sparsely punctate; head and antennae densely
punctate. Punctation of thorax, abdomen, and legs similar to that of head
but much more shallow.

Head and body with silky pubescence which is most abundant on head and
gaster and lacking on petiolar node; ever^-where with sparse, very fine erect
pile.

Color brownish yellow; vertex with a small fuscous spot; and tarsi and
mandibles a little darker. ,

San Cristoval: Wainoni Bay.

Described from one worker.

Related to E. (T.) crassicornis Emery, known only from a female
from New Guinea, but the petiole is much longer than deep and the
thorax is not shining and the funicular articles 2-4 are not much
broader than long.

Dedicated to Frederick Sheldon, in whose memory the Sheldon
Travelling Fellowships of Harvard University were established.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 293

19. Cryptopone mayri, sp. nov.

Worker. Length 2.5-2.75 mm.

Head a little longer than broad, slightly narrowed in front, with moderately
convex sides, broadly rounded occipital corners and shallowly concave border.
Base of cl5qDeus carinate at middle, anterior border feebly rounded. Front
with an acute median carina. Mandibular blades with five stout triangular
teeth. Antennae stout, their scapes thickened distally, extending nearly
four fifths the distance to occipital corners; club distinctly longer than re-
mainder of funiculus, middle joints strongly transverse; terminal joint longer
than the two preceding joints together. Eyes absent. Prothorax slightly
convex above, submargined in front and at sides. Mesothorax flattened,
transverse. Promesonotal and mesoepinotal sutures distinctly impressed.
Base of epinotum flat, about as long as the declivity, which is flat and margined
above at sides. Petiolar node thick, two thirds as long as high, anterior sur-
face slightly concave, and narrowly margined at sides; upper surface broadly
rounded; posterior sm-face rather flat, evenly rounding into the dorsal surface.
Gaster long and slender.

Mandibles shining, sparsely punctate. Head, prothorax, and antennae
densely punctate and opaque. Mesothorax, epinotmn, gaster, and legs
equally densely but more shallowly punctate and somewhat shining.

Pruinose pubescence and sparse erect pile on head and body.

Yellowish brown; head fuscous.

Female. Length 3 mm.

Similar to worker. Eyes large and flat, situated at a distance equal to half
their length from front of head. Ocelli small. Wings (length 3 mm.) strongly
infuscated.

Ysabel: Fulakora (Type-locaUty). L"gi: Pawa. San Cristoval:
Wai-ai, Wainoni Bay, Pamua.

Occurs in small colonies beneath stones. In the absence of eyes
C. mayri is distinct from the three described Papuasian species and
more closely related to iestacea Motsch. from Ceylon. It differs from
that species in having the head longer, with the sides much less con-
vex, judging from Emery's figure (Ann. Soc. ent. France, 1893, 62,
pi. 6, fig. 3). Type. — M. C. Z. 9,156.

C. fusciceps Emer}', besides having distinct though minute eyes, is
smaller in size, but is otherwise similar in habitus to mayri, and evi-
dently resembles the following variety.

294 bulletin: museum of comparative zoology.

19a. Cryptopone mayri Mann, var. fuscior, var. nov.
Worker. Length 2 mm.

Differing from the preceding only in its smaller size and in the color,
being dark fuscous almost black with the anterior border of head,
mandibles, borders of gastric segments, and appendages brown.

Ysabel: Fulakora.

Described from two workers.

Key to Papuasian Species.

Eyes absent 1 .

Eyes present, though minute 2.

1. Length 2.50-2.75 mm. Color yellowish brown. (Solomons) mayri Mann.
Length 2 mm. Color mostly black. (Solomons) . . mayri var. minor Mann.

2. Mesgepinotal suture distinct, mandibles 5-dentate. Length If-l* mm.

(New Guinea) fusciceps Emery.

Mesoepinotal suture indistinct; head longer. Mandibles with three to
four strong teeth , 3.

3. Antennal scapes extending less than two thirds the distance to occipital

corners; mandibles with three large teeth anteriorly and obtusely
dentate posteriorly. Length 1-L2 mm. (New Guinea) . . mocsaryi Czabo.
Antennal scapes extending more than two thirds the distance to occipital
corners; head a fourth longer than broad; mandibles with four teeth in
front, edentate behind. Length If-lf mm. (New Guinea).

tenuis Emery.

20. Ponera gleadowi Forel, subsp.

San Cristoval: Pamua.

A unique worker belongs to a subspecies of gleadowi near, if not iden-
tical with subsp. decipien,, Forel from Hawaii.

21. Ponera pruinosa Emery.

Term, fuzet., 1900, 23, p. 316, 319, pi. 8, fig. 13, 14, g .

Female. Length 3.5 mm.

Scarcely larger than the worker. The eyes are a little more than a fourth
as long as the head, situated at a distance equal to half their length from the

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 295

clypeus. The wings (length 3.25 mm.) are faintly infuscated, and densely
pubescent; veins and stigma fuscous.

Male. Length 2| mm.

Head, excluding the eyes, as broad as long, broadly rounded behind. Man-
dibles feeble, short, triangular, and edentate. Clypeus strongly elevated at
middle, the anterior border narrowly concave at middle. Antennae long and
slender; first funicular joint scarcely longer than broad, a little more than half
the length of the scape; 2nd joint two and three fourths times as long as the
first; joints 3-11 subequal, cylindrical; terminal joint one and a half times as
long as penultimate. Eyes and ocelli large, the former moderately convex.
Epinotum broad; in profile convex basally; declivous portion sloping, with
the surface flat discally, broadly rounded at sides. Node shorter than that
of worker; anterior face convex, rounded above, declivous behind. Gaster
markedly constricted between first and second segments. Genitalia small.

Color and pubescence similar to that of worker; pilosity more abundant.
Wings not infuscated, pubescent similarly to those of female.

Three Sisters: Malapaina. San Cristoval: Wainoni Bay, Wai-ai,
Pamua. Ugi: Pawa. Malaita: Auki. Florida: Tulagi. Ysabel:
Fulakora.

A large series of workers agree closely with Emery's description of
workers from New Guinea. It is the commonest species of the genus
in the Solomons.

22, PoNERA PAPUANUM Emery.

Term, fuzet., 1900, 23, p. 319, pi. 8, fig. 10-11, ^ .

Three Sisters: Malapaina. San Cristoval: Wai-ai. Ugi: Pawa.
Ysabel : Fulakora.

A small series of workers and females which I refer to this species
agree closely with Emery's description.

The wings of the female are strongly infuscated, with the veins and
stigma dark.

23. PoNERA PALLiDULA Emery.

Term, fuzet., 1900, 23, p. 316, 320, pi. 8, fig. 17, 18, ^ .

Malaita: Auki. Ysabel: Fulakora.

A dealated female taken in a colony is barely larger than the worker.

296

bulletin: museum of comparative zoology.

24. PoNERA CLAVicoRNis Emerv.

Term, fuzet., 1900, 23, p. 317, pi. 8, fig. 7, 8, g .
Ysabel : Fulakora.

25. Leptogenys (Leptogenys) truncatus, sp. nov.
Worker. Length 6.5 mm.

Related to L. emeryi Forel. Head slightly longer than broad, occipital
angles broadly rounded, border straight. Frontal carinae short, their lobes
small. Clypeus strongly and acutely carinate at middle, projecting in front,
the projected portion twice as broad as long with a feebly biconvex border.

Mandibles shorter than sides of head.
Antennal scapes slightly surpassing the
occipital corners; funicular joints one
and three subequal in length and shorter
than the second; joints four to ten sub-
equal; terminal joint a little less than
twice the length of penultimate. Pro-
thorax broader than long; rounded in
front and at sides; in profile only slightly
convex above. Mesothorax a little
broader than long, the sides and front
rounded. Mesoepinotal impression pro-
found. Epinotum convex, broadest be-
hind; in profile the basal portion is one
and two third times as long as the de-
clivity. Petiole very slightly longer than
broad; in profile, slightly higher than
long, rounded in front and at top; pos-
terior surface flat. Gaster slender; con-
striction between first and second gastric segments not strong.

Shining. Front rugulosely punctate, vertex at middle with short trans-
verse and reticulate carinulae; remainder of head, thorax, epinotum (except
declivity), and petiole with coarse foveolate punctures which are more dense
and confluent on the epinotum than elsewhere. Epinotal declivity trans-
versely striate. Gaster more shallowly punctate. Mandibles and antennae
shining, punctate.

Pilosity long and fine, abundant; erect on head and body, semierect on
antennae. The femora and tibiae have, in addition to semierect pile, finer and
recumbent pilosity.

Fig. 9. — Leptogenys (Leptogenys) trun-
catus Mann. Worber. Head showing
epinotum and clypeus.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

297

Color black; legs and antennae lighter; mandibles red.
white.

Pilosity yellowish

Santa Cruz: Graciosa Bay.

The description is based on a single worker. This species is closely
related to cmcryi Forel from the Bismarck Archipelago, but the clypeus
is entirely different, with the middle projection bisinuate instead of
trisinuate.

26. Leptogenys (Leptogenys) foreli, sp. nov.

Worker. Length 6.5-7 mm.

Head a little broader than long, broadest in front, posterior border straight.
Mandibles about as long as sides of head, strongly curved basally and then
slightly arcuate and of even thickness to
tips. Clypeus sharply carinate at middle;
median lobe longer than broad, with sides
parallel for half their length, then slightly
concave and converging to form a triang-
ular pointed tip; lobes on either side of
middle broadly rounded. Antennal scapes
surpassing occipital corners by about one
third of their length; first and third joints
subequal; terminal joint shorter than the
two preceding joints together. Thorax
broadly and shallowly impressed between
meso- and epinotum. Base of epinotum
convex above, much longer than the de-
clivity (in emeryi Forel but little longer)
and rounding into it. Petiolar node slightly
broader than long, convex above, highest
behind, with straight posterior and nearly straight anterior surfaces.

Head and thorax subopaque. Mandibles striate. Head rugose and with
coarse foveolate punctures scattered on front and occiput. Pro- and mesono-
tum coarsely, foveolately punctate, the interspaces finely rugulose and more
shining than the rest. Epinotal declivity transversely striate, the base and
the petiolar node rugose. Gaster shining, with two sizes of punctures. Scapes
and legs finely punctate.

Long erect pilosity abundant everywhere.

Black; mandibles, antennae, apex of gaster, tibiae, and tarsi reddish brown.

Male. Length 6.5 mm.

Head a httle longer than broad, broadly and evenly rounded behind.
Mandibles small and spatulate. Clypeus with an elongate tubercle a little

Fig. 10. — Leptogenys (Leptogenys)
foreli Mann. Worker. Head show-
ing epinotum and clypeus.

298 bulletin: museum of comparative zoology.

in front of middle; anterior- border feebly arcuate at middle. Eyes and
ocelli large. Antennae slender; scape twice as long as first funicular joint,
which is about as broad as long; remaining joints becoming very gradually
shorter toward apex; terminal joint one and one third times as long as the
preceding joint, the apical third of it conical in shape. Prothorax in profile
a third as long as mesothorax. Mesothorax flattened distally, with strong
Mayrian furrows. Scutellum strongly convex, a httle broader than long.
Epinotum rounded above, disc of the declivous portion fiat. Petiolar node
in profile as long as high, anterior face convex and broadly rounding into the
top, posterior surface flat; from above, as long as broad, rounded in front and
at sides, truncate behind. A strong constriction between first and second
gastric segments.

Somewhat shining. Head and prothorax striolate rugulose. Mesothorax
and scutellum wath similar sculpture and strongly and foveolately punctate.
Epinotum coarsely rugose. Petiolar nodes coarsely striate at sides and
heavily punctate throughout. Gaster finely punctate.

Head and legs with long and silky recumbent hairs. Pilosity abundant;
on the gaster longer and less erect.

Color black; mandibles, antennae, geniculae, and tarsi brown. Wings
inf uscated ; veins and stigma brown.

Three Sisters: Malapaina (Type-locality). Malaita: Auki, Simoli
(Coll. H. Hall).

Closely related to emcryi Forel, from the Bismarck Archipelago,
from which it differs in the structure of the epinotum and the clypeus.

Key io the Papuasian Species of the Subgenus Leptogenys.

1. Length 12 mm. Head distinctly longer than broad, with scattered pili-

gerous punctures. Postpetiole and gaster rufous. (New Guinea).

triloba Emery.

Length 5 to 8 mm. Head not or but slightly longer than broad, rugosely

punctate. Gaster black, sometimes with bluish reflections, rufous

apically 2.

2. Mandibles distinctly longer than sides of head. Petiole much longer

than broad. Lobe of cl3rpeus elongate and rounded apically. Length

7.5-8 mm. (Engano) modiglianii Emery.

Mandibles not longer than sides of head. Petiole at least as broad as
long 3.

3. Lobe of clypeus broad and trisinuate in front. Base of epinotum but little

longer than the declivity. Length 5.8 mm. (Bismarck Archipelago).

emeryi Forel.

Lobe of clypeus not trisinuate in front. Base of epinotum much longer

than the declivitj^ 4.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 299

4. Lobe of cljT^eus broader than long, broad in front and slightly concave at

middle of border. Length 6.5 mm. (Santa Cruz). . . .truncata Mann.

Lobe of cljT^eus longer than broad and pointed apically. Length 6.5-

7 mm. (Solomons) foreli Mann.

27a. Leptogenys (Lobopelta)diminuta(F. Smith), subsp. santschi,

subsp. nov.

Worker. Length 6.25 mm.

Differing from the tjqjical diminuta in its somewhat more slender habitus
and in the following characters: — the mesoepinotal impression is narrower.
The mesopleurae are strongly, densely, and accurately striate. The epinotal
base is not as rugose as the declivity, which has regular, transverse, and parallel
striae, stronger than in diminuta and not interrupted by the lateral tubercles.
The latter are much smaller than in diminuta so the declivous surface is broader
at the base. The petiolar node is higher in proportion to the length.

The front of head has arcuate striae as in diminuta. The body, node, and
gaster are shining.

Female. Length 7.5 mm.

One specimen among a very large series, has the gaster longer and miore
enlarged than the others and evidently represents the sexual phase. The
petiole is very slightly thinner in profile than in the ordinary workers, but
otherwise there is no difference in structure.

Male. Length 5 mm.

Head, excluding eyes, longer than broad. ALindibles short and bluntly
pointed. Clypeus large and convex, the anterior border slightly concave at
middle. Antennal scape more than twice as long as the first funicular joint;
second funicular joint three and one half times as long as the first; succeeding
joints gradually becoming shorter; terminal joint one and five eighths times
as long as the preceding. Eyes and ocelli large and prominent. Prothorax
little convfex above. Scutellum longer than broad. Epinotum convex basally,
the declivous portion flat. Petiolar node in profile longer than high; seen
from above, circular in shape. First and second gastric segments with a
strong constriction between.

Feebly shining; head punctate and finely striolate; the striolae surrounding
the ocelli coarser than elsewhere. Thorax rugulosely striate, the striae on
apical half of mesothorax converging toward the center; those on scutellum
longitudinal and finer. Epinotum granulosely punctate. Petiolar node and
gaster finely and sparsely punctate and shining. Femora densely and rather
coarsely punctate. Clypeus and legs with semierect pile.

300 bulletin: museum of comparative zoology.

Color fuscous; with mandibles, antennae, and geniculae lighter. Wings
(length 4 mm.) hyaline, with semierect and moderately long hairs; veins and
stigma pale.

Three Sisters: Malapaina (Type-locality). San Cristoval: Wai-ai,
Pamua. Ugi: Pawa.

Type. — M. C. Z. 9,157. On several occasions I found masses of
workers, accompanied by males, swarming on the ground, always in
the forest, and numerous larvae and pupae beneath pieces of bark
lying on the ground. Probably these were temporary nesting places.

This form was found only in the eastern end of the Solomons.

27b. Leptogenys (Lobopelta) diminuta (F. Smith), var. laeviceps

(F. Smith).

Ponera laeviceps F. Smith, Journ. proc. Linn. soc. London. Zool., 1858, 2,
p. 69, S .

Malaita: near Fourafi, in the interior.

A small series of workers taken running in file oh the ground,
across a trail, evidently belong to this variety.

28. Anochetus graeffei Mayr.
Verh. K. K. zool.-bot. gesellsch. Wien, 1870, 20, p. 961, ^ .

Santa Cruz: Graciosa Bay.

One dead and broken worker of this widely distributed Polynesian
species was found.

29a. Anochetus punctiventris Mayr, subsp. oceanicus Emery.
Ann. Mus. civ. stor. nat. Genova., 1884, 21, p. 378, ^ (nee. Mayr).

San Cristoval: Wainoni Bay, Wai-ai, Pamua. Malaita: Auki.
Florida: Tulagi. New Georgia: Maravo Lagoon.

There is some variation among workers from the same colony in
the shape of the petiole, which in some specimens is more rounded
above than in others. Three females among my specimens have the
upper border distinctly emarginate at the center.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 301

30. Anochetus CATC Forel.

Mitt. Mus. zool. Berlin, 1901, 2, heft 1, p. 6, ^ .

Rendova.

The workers of a single colony taken from a rotten log, agree
throughout with Forel's description of this species, which is character-
ized by its rounded epinotal corners, the short conical projection at the
apex of the petiolar node and the structure of the mandibles which are
sharply angulate at a short distance posterior to the long, terminal
teeth.

The sculpture on the front of head is dense, and extends outward
from the frontal carinae.

30a. Anochetus cato Forel, var. subfasciatus, var. nov.

Worker. Length 5.-5.25 mm.

Similar to the typical form in size, structure, and sculpture but black in
color, with the cheeks, clypeus, mandibles, legs, and a narrow apical band on
each gastric segment, brown.

Female. Length 7.25 mm.

Epinotumstrongly striate transversely; eyes not large ; ocelli small. Wings
weakly infuscated; veins and stigma brown. Otherwise similar to the worker.

Male. Length 4 mm.

Head, excluding eyes as broad as long; the medial portion of vertex on each
side with longitudinal sulcae which diverge and extend.to the eyes. Mandibles
rudimentary. Antennae 13-jointed, rather thick; scape less than twice as
long as first flagellar joint, which is a little longer than broad; flagellar joints
two to six subequal, only a little more than twice as long as broad; remaining
joints, except the terminal, subequal; terminal joint shorter than the two
preceding together; conical. Prothorax evenly rounded at front and sides,
its disc not very convex. Scutellum transverse. Epinotum convex; the
base rounding into the short declivity. Petiolar node, in profile, wedge-
shaped; thin and narrow above; seen from the front, not produced above,
but evenly rounded; anterior surface concave; narrowly margined at sides.
Pygidium acuminate apically.

Somewhat shining. Head and thorax evenly, rather coarsely punctate.
Epinotum rugose. Node and gaster more sparsely punctate.

302

bulletin: museum of comparative zoology.

Finely pilose throughout; antennae densely covered with short stiff re-
cumbent hairs.

Color black. Wings faintly infuscated; veins and stigma brown.

Three Sisters : Malapaina (Type-locality). San Cristoval : Wai-ai.
Ugi: Pawa. Malaita: Auki. Florida: Tulagi. Ysabel: Fulakora.
Type.—M. C. Z. 9,158.

30b. Anochetus cato Forel, isolatus, subsp. nov.

Worker. Length 6.5 mm.

Differing from A . cato Forel in the following characters : — the petiolar
node in profile is thicker, more conical and blunter at apex; the anterior sur-
face is evenly convex and not constricted before the apex as in cato.

The striation on the front of head is feebler and sparser and extends barely
past the ends of the frontal carinae and is included between them.

The head, thorax, and epinotum are black, the gaster, legs, mandibles, and
antennae, brownish red.

Fig. 11. — Anochetus cat© isolatus Mann. Worker. Lateral view of thorax and petiole.

Male. Length 5 mm.

In addition to its much larger size, differing from the male of cato in the
structure of the petiolar node, which in profile is nearly as broad as long and
much less narrowed above.

The wings are less infuscated than in cato.

Santa Cruz: Graciosa Bay (Type- locality). Three Sisters:
Malapaina.

Several small colonies were found in the first named locality and
solitary workers in the last. Type. — M. C. Z. 9,159.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 303

Key to the Species {Workers) from New Guinea and the Solomons.

Mandibular blades at f distance from apex to base with an erect and acute
tooth; epinotum bidentate. Length 41-5 mm. (New Guinea).

chirichmii Emery.

Mandibular blades with acute teeth at the middle; epinotum angulate
or rounded 1.

1. Petiolar node strongly transverse; in profile narrow and wedge shaped;

seen from front rounded, truncate or concave above; epinotum angulate

between base and declivity. Length 3.5-5 mm 2.

Petiolar node in profile conical or subcorneal; epinotum rounded between
base and declivity. Length 6-7 mm 3.

2. Basal gastric segment not coarsely punctate graeffei Mayr.

Basal gastric segment coarsely punctate.

punctiventris Mayr subsp. oceanicus Emery.

3. Striation on front of head not dense; extending only slightly beyond ends

of frontal carinae and confined between them; bicolored. (Santa Cruz :

eastern Solomons) cato Forel subsp. isolatus Mann.

Striation on front of head dense, covering most of the front, and extending
two tliirds the distance to occipital border and not confined between
frontal carinae 4.

4. Ferruginous (New Britain and western Solomons) cato Forel.

Mostly black (Solomons) cato Forel var. subfasciatus Mann.

3L Odontomachus haematoda (Linne).

Syst. nat., ed. 10, 1758, 1, p. 582, ^ .

Santa Cruz: Graciosa Bay. San Cristoval: Parana, Wainoni Bay.
Three Sisters: Malapaina. Ugi: PaAva. Russell: Yandina.
Malaita: Auki, Simoli (Coll. H. Hall). Florida: Tulagi, Maliali.
Ysabel : Fulakora.

As abundant in the Solomons as it is in all other tropical countries.

32a. Odontomachus imperator Emery, subsp. emeryi, subsp. nov.

Worker. Length 14.5-15.5 mm.

Head, and gaster black, mesothorax dark brown; prothorax, epinotum,
petiole, mandibles, antennae, and legs yellowish brown, the tarsi darker than
the other parts.

304

bulletin: museum of comparative zoology.

Mandibles densely and finely striolate and with elongate punctures. Front
of head between frontal carinae longitudinally striate; remainder of front
except between eyes and antennae, strongly and concentrically striate; lateral
fossae at the anterior border with short striae, but most of the surface smooth
and shining; vertex and sides densely striate, the striae becoming more subtle
posteriorly and lacking for a short distance from the sides of the medial impres-
sion; occiput sparsely punctate and shining. Prothorax transversely and
somewhat arcuately striate; meso- and epinotum transversely striate. Base
of petiolar node striate. Gaster finely punctate and shining.

Female. Length 19 mm.

Mesonotum and scutellum with strong longitudinal striae.

Anterior portion of prothorax, a discal spot, disc of mesothorax and anterior
face of scutellum fuscous. The rest as in worker, with the usual sexual
differences.

Wing (length 12 mm.) faintly infuscated; veins and stigma brown.

Male. Length 12 mm.

Head, excluding eyes, longer than broad, broadly rounded behind. Mandi-
bles rudimentary. Cl3rpeus convex, the anterior border almost straight.

Antennal scape thick, three times
as long as the first funicular joint;
funicular joints, except the first,
very long and slender; the second
three times as long as the scape;
remaining joints gradually shorter;
with the terminal joint less than
twice as long as the penultimate.
Eyes emarginate on inner border.
Ocelli large and prominent. Me-
sonotum little convex above. Scu-
tellum a little broader than long,
somewhat pyramidal in shape,
with a short, longitudinal impres-
sion at apex. Epinotum slightly
convex basally, with a well-marked
declivous portion. Petiole two
and a half times as long as broad;
in profile twice as long as high; anterior face concave, narrowly margined at
sides; upper surface elevated into a flat projection that is strongly and sharply
margined on its posterior face; spiracles on sides situated at tips of strong,
elongate tubercles. Gaster long and slender.

Subopaque. Head, pro- and mesothorax, and scutellum delicately striolate
and punctate. Epinotum more strongly, obliquely longitudinally striate.

Fig. 12. — Odontomachus imperator emery i
Mann. Males. Lateral views showing varia-
tions in petiole.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

305

Petiole densely and microscopically punctate. Gaster shining and finely
punctate.

Head, thorax, and gaster with short and subappressed golden pubescence
and a few longer erect hairs.

Color yellow. Wings faintly yellowish; veins pale.

Florida: Maliali (Type-locality). Ysabel: Fulakora.

Differing from iviperator subsp. rufithorax Emery in color, not having
the thorax and epinotum " rouge sanguin," and in the sculpture of the
head; and from subsp. opacuhis Viehmeyer, in color.

The yellow male is very Ichneumon-like. The two specimens,
taken from the same colony, show a marked difference in the struc-
ture of the petiolar node. In one the dorsal surface gradually slopes
to and up the sides of the triangular projection ; in the other there is
an angle between the two. The lateral tubercles on the node are
unusually large.

Several colonies were observed. They were in dense forest; the
nests were in the ground among the roots of trees and contained large
numbers of workers. The workers are less active than haematoda and
not as aggressive. Type. — M. C. Z. 9,160.

33a. Odontomachus malignus F. Smith, subsp. tuberculatus

Roger.

Berl. zeitsch., 1861, 5, p. 28, g .

Santa Cruz: Graciosa Bay. Malaita:
Simoli (Coll. H. Hall).

A large series of workers agree with
Roger's description of tuberculatus in
having the mesonotum longitudinally
striate. Smith described inaUcjnus as
having the mesonotum transversely stri-
ate, so this is evidently a distinct form of
at least subspecific value.

It is a beautiful ant, bright ferruginous
in color and distinct from other species
of Odontomachus in having the vertex
bituberculate. The head is constricted
behind, but is proportionately smaller
and shorter than in other species belong-
ing to the hastatus group.

I found the species only once, at ^'''- 13.— Odontomachus malignua
r^ • r, 1 \ tuberculatus Roger. Worker.-

(jrraciosa liay, where workers were Front view of head.

306 bulletin: museum of comparative zoology.

moving in and out of the crevices of a large block of coral on the beach.
Mr. Harry Hall, who brought me specimens from Simoli on South
Malaita, states that he found it there nesting under the same condi-
tions.

Myrmicinae.

34. SiMA (Tetraponera) humerosa Emery.

Ann. Mus. civ. stor. nat. Genova, 1900, 40, p. 674, y .
Ysabei : Fulakora.

35. Pheidole (Pheidolacanthinus) belli, sp. nov.

Worker. Length 2.25 mm.

Head about as long as broad, with convex sides and nearly straight occipital
border. Mandibles with ten small teeth. Clypeus convex; anterior border
broadly rounded. Antennal scapes surpassing occipital corners by about
one fifth their length; first funicular joint longer than the two succeeding
joints together; joints two to eight a little longer than broad; club slender,
slightly shorter than the remainder of funiculus. Eyes small, convex, situated
a little in front of middle of sides of head. Pronotum flattened at middle,

Fig. 14. — Pheidole (Pheidolacanthus) belli Mann. Worker.

armed with a pair of very long and slender spines that extend outward and
forward and are rather strongly cm"ved downward at tips. Base of epinotum
as long as declivity, flat; epinotal spines as long as those of pronotum, ex-
tending outward and backward and rather strongly curved downward at tips.
Petiole long and slender; node in profile triangular, gradually sloping into the
pedicle; from above, as long as broad, with the sides subconulate. Post-
petiole longer than broad, with slightly convex sides. Gaster long and rather
narrow. Legs long and slender.

mann: ants of the British solomon islands. 307

Shining throughout, and smooth, except for very fine scattered punctures
on head and thorax and delicate rugulae on apical part of mesonotum; base
of first gastric segment with very broad and shallow foveolate punctures.

Sparse, long, erect pile scattered on head, body, and appendages.

Black; mandibles, antennae, and legs dark reddish brown. Pilosity dark.

Malaita: near Fourafi, in the interior.

Described from one worker.

This species, w^hich is dedicated to Mr. W. P. Bell, Resident Com-
missioner of Malaita, resembles P. flavoihoracica Viehmeyer but
differs in having the thoracic and epinotal spines more slender and
strongly curved, in the structure of the petiole and in color.

I follow Viehmeyer in considering the above species as belonging
to the subgenus Pheidolacanthinus, although it has 12- jointed
antennae. From Smith's description and figure of the genotype
(P. armaim) that form is merely a Pheidole allied to the better known
P. sexspinosa Mayr. and belonging to a group of species with armed
pronota that may be considered of subgeneric rank.^

36. Pheidole (Pheidolacanthinus) sexspinosa Mayr.

Verh. KK. zool.-bot. gesellsch.Wien, 1870, 20, p. 977, ^ %.

Santa Cruz: Graciosa Bay. San Cristoval: Wai-ai, Wainoni Bay.
Malaita: Auki. Ysabel: Fulakora.

36a. Pheidole (Pheidolacanthinus) sexspinosa Mayr, var.

FUSCESCENS Emery.

Term, fuzet., 1900, 23, p. 32.3, ^ %.
Florida: Tulagi.

37. Pheidole (Pheidolacanthinus) erato, sp. nov.
Soldier. Length 3.5 mm.

Head large, nearly twice as broad as pronotum, a little longer than broad,
with slightly convex sides, rounded occipital corners, and narrowly and deeply

1 In response to my request to examine the type of Pheidolacanthinus, Mr. Horace St. J.
Donisthorpe writes as follows: —

"Smith's type of Pheidolacanthinus armalas is not in the British Museum. As however it is
at Oxford, I have got Prof. Poulton to examine it for me and he reports that the antennae are
12-jointed."

308 bulletin: museum of comparative zoology.

impressed border; transverse impression of vertex strong. Mandibles stout,
bluntly dentate apically; outer border a little concave in front of middle.
Clypeus flat, its borders poorly defined, carinate, concave at middle of anterior
border. Frontal area deeply impressed. Frontal carinae divergent, as long
as the scape and bordering a moderately well-defined scrobe. Antennal
scapes extending a little more than half the distance to occipital corners.
Club slender, funicular joints longer than broad. Eyes small and convex,
situated a little in front of anterior third of sides.

Promesonotum convex. Pronotmn with a pair of spines pointing upward
and outward, about two thirds as long as width of pronotum, thick basally and
acuminate at tips. Mesonotum transversely ridged and with two small
denticles. Epinotum with a narrow margin between basal and declivous
surfaces; both surfaces concave; spines twice as long as their distance apart
at base, only slightly divergent and curving a little backward, angulately
denticulate at anterior third. Petiolar node with flat anterior and concave
and margined posterior surfaces; impressed at upper border, sides elevated
and bluntly triangular. Postpetiole a little broader than long; broadest in
front, with rather prominent, bluntly angulate anterior corners.

Moderately shining, except gaster, which is subopaque. Mandibles shining,
with coarse scattered punctures. Sides of clypeus striate. Head with coarse
ii'regular striae extending as far as tips of scapes, intervening spaces rugulose,
vertex, occiput, pronotum, and mesonotum very coarsely rugose-striate.
Scrobe granulose. Base of epinotum finely and transversely striate. Post-
petiole with broad, longitudinal sulci. First gastric segment very densely
striate and punctate tlu-oughout; granulose basally; second and third seg-
ments densely punctate. Legs smooth and shining.

Scattered, erect, yellow hairs present on head, body, and appendages.

Color brownish red, mandibles red. Legs yellow.

Worker. Length 2 mm.

Head as long as broad, rounded at sides and behind. Mandibles 7-dentate .
Clypeus flat, broadly rounded anteriorly. Antennal scapes surpassing occipi-
tal corners by more than one fourth their length; funicular joints two to seven
about as broad as long. Pronotum flat, pronotal spines slender, as long as
their distance apart at base, directed forward and outward and slightly curved
downward. Epinotum concave between the spines, which are twice as long
as distant apart at base, erect and rather strongly curved backward. Petiole
four times as long as broad; node feebly margined in front and sides. Post-
petiole longer than broad, broadest behind, with slightly convex sides.

Mandibles shining. Head subopaque, granulosely punctate with a few
fine longitudinal striae on front and cheeks. Promesonotum punctate simi-
larly to head, but more shallowly, with striae more abundant and more shining.
Epinotum subtly punctate and shining. Petiole, postpetiole, gaster, and legs
shining.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 309

Very fine, yellow pile scattered throughout.
Uniformly yellow.

Female (dealated). Length 4.50 mm.

Head about as broad as long; moderately concave behind at middle. Fur-
row of vertex shallow. Front with a deep pit anterior to median ocellus.
Eyes small, situated at sides in front of middle. Pronotum with short, stout,
triangular spines. Epinotal spines a little shorter than their distance apart
at base, stout. Petiole shorter than in the soldier; node similar to that of
soldier but with the lateral angles more rounded. Postpetiole twice as broad
as long, broadest in front.

Sculpture of head similar to that of soldier; scrobe coarsely granulose.
Mesothorax longitudinally rugosely striate. Scutellum rugose. Epinotum
with fine oblique striae. Petiole, postpetiole, and gaster with sculpture
similar to that of soldier.

Pilosity as in soldier.

Color dark ferruginous, with the legs paler.

Ysabel : Fulakora. '

Described from a single soldier, a female, and two workers.
The almost opaque gaster, with the second and third, as well as the
first segment, densely punctate, and the color separate erato from
tetracaniha Emer3\ Judging from Emery's description teiracaniha
has shorter epinotal spines. The two forms are closely related; the
difference may be only subspecific. P. singularis Smith has the
gaster similarly punctate, but has shorter antennal scapes and entirely
different frontal carinae, as shown in Emery's figure (x\nn. Mus. civ.
stor. nat. Genova,1887, 25, pi. 1, fig. 13).

38. Pheidole mendanai, sp. nov.

Soldier. Length 4.5-5 mm.

Head large, longer than broad, with nearly straight sides and deeply e.x-
cised border; transversely depressed at vertex; frontal sinus profound to a
point a little behind the level of eyes where there is a foveolate pit and more
feebly impressed anterior to this, extending to a point opposite the anterior
border of eyes and terminating in a small pit. Mandibles stout, very bluntly
bidentate at tip. Clypeus triangular, the surface depressed and flat, weakly
carinate for entire length, anterior border depressed and concave at middle.
Frontal area not distinct. Frontal carinae shorter than the scapes, dilated

310

bulletin: museum of comparative zoology.

basally, rather broadly covering the antennal insertions; behind no stronger
than the other carinae on front. Antennae short and slender, their scapes
extending less than half the distance to occipital corners; club slender, shorter

than remainder of funiculus; joints two to eight
distinctly longer than broad. Eyes small (about
thirty facets), convex, situated at anterior fourth
of sides. Pronotum rounded, sides produced
posteriorly into short blunt cones, which have
the borders distinctly margined. Promesonotal
suture feebly impressed. Mesonotum submar-
gined at sides, flat above; behind descending
abruptly to the deeply impressed mesoepinotal
suture. Epinotum concave on basal and decli-
vous surfaces, armed with stout, erect spines
which are about as long as distant at base.
Petiole a little more than twice as long as broad,
flattened above, margined at sides; node from
above four times as broad as long, submargined
behind and somewhat impressed at middle.
Postpetiole twice as long as broad, rounded in
front and behind, submargined above; sides produced into coarse blunt
spines, each about half as long as the width of petiolar node.

Moderately shining. Mandibles delicately striolate and coarsely, regularly
punctate. Clypeus smooth, except for two striae laterally. Front and cheeks
longitudinally striate, the striae dense oij cheeks and more widely separated
on front, with the intervening spaces delicately rugulose. Vertex and occiput
reticulately striate and rugulose, the striae strongest on occiput. Thorax
and base of epinotum sparsely, delicately, and irregularly striated transversely,
Node of petiole and postpetiole rugulose. First gastric segment densely
striolate throughout.
Legs shining.

Fine silky erect hairs present on head, body, and appendages, shorter and
finer on head.

Ferruginous; gaster lighter.
legs pale.

Fig. 15. — Pheidole mendanai
Mann. Soldier. Front view
of head.

Mandibles and anterior border of head darker.

Worker. Length 2.25 mm.

Head a little longer than broad, as broad in front as behind, sides slightly
convex, posterior border broadly and shallowly concave. Mandibles elongate,
with five fine and two larger, triangular teeth. Clypeus flattened anteriorly
and carinate at middle of anterior two thirds; anterior border nearly straight.
Antennal scapes surpassing occipital corners of head by a little less than a
third of their length; funicular joints longer than broad; club slender, shorter
than remainder of funiculus. Eyes convex, situated a little in front of middle

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 311

of sides of head. Pronotum convex, sides at posterior third produced into
short, margined angles. Mesonotum fiat, sloping toward mesoepinotal suture.
Base of epinotum flat, submargined at sides, concave between the spines,
which are stout and as long as their distance apart at base; basal portion con-
cave and feebly margined laterally. Petiole nearly four times as long as broad ;
node about three times as long as broad, rounded at sides. Postpetiole one
and a third times as broad as petiole, rounded above and at sides.

Shining. Mandibles coarsely punctate. Cheeks with regular, coarse,
separated striae; front with fine striae; vertex and occiput rugulosely punctate
and delicately reticulately striate. Thorax, petiole, and postpetiole very
finely rugulose. First gastric segment minutely punctate.

Fine erect hairs moderately abundant on head, body, and appendages.

Testaceous to light ferruginous.

Female. Length 6 mm.

Head barely broader than long, narrowed anteriorly, concave at occipital
border, with a furrow extending to the median ocellus. Mandibles and clypeus
similar to those of soldier. Antennal scapes extending to lateral ocelli.
Thorax narrower than the head. Epinotal spines strong, slightly shorter
than their distance apart at base. Petiole more narrowed above than in the
soldier and more distinctly margined and more strongly depressed at middle.
Postpetiole similar to that of soldier.

Head striate on front and cheeks similar to soldier, but the striae of vertex
and occiput are less reticulate and the spaces between are not as rugose.
Mesothorax strongly longitudinally striate. Scutellum at basal half smooth,
except for a pair of foveolate punctures; at apical half irregularly striate and
rugulose. Petiole, postpetiole, and first gastric segment sculptured as in
worker.

Hairs yellowish, long and moderately abundant.

Ferruginous; wings infuscated.

Ysabel : Fulakora.

Described from specimens taken from a colony found beneath a log.
r?//;e.— M.C.Z. 9,161.

The closely related P. laminata Emery from New Guinea differs in
having the frontal lamellae produced more angularly in front and in
its much smaller size. In mendanai, as in laminata and cryptoccra
Emery the gaster is microscopically scabrose and subopaque.

39. Pheidole ISIS, sp. nov.

Soldier. 2.75-3 mm.

Head a fourth longer than broad, sides parallel; posterior border deeply
concave at middle and vertex very strongly impressed transversely, so that

312

bulletin: museum of comparative zoology.

the posterior corners stand out prominently, somewhat lobiform. Mandibles
thick, outer border straight at middle, blades edentate. Clypeus flat, concave
at anterior border, strongly carinate. Frontal area deeply impressed. Frontal

Fig. 16. — Pheidole isis Mann. Soldier. Lateral view of head, thorax, and petiole.

lamellae flat, dilated; frontal carinae strong, expanded laterally, a little longer
than the scapes, and bordering a broad scrobe capable of containing the entire
scape (similar to cryptocera Emery). Antennae short, their scapes extending
less than half the distance to occipital corners; funicular joints two to eight
strongly transverse; club moderate, shorter than remainder of funiculus.
Eyes small (about fifteen facets) and convex, situated at sides of head well in
front of middle. Promesonotmn without suture, about as broad as long, sides

strongly produced conically and ending in stout tri-
angular spines; sides of mesothorax obtusely angulate,
posterior portion perpendicular to the mesoepinotal
impression. Epinotum concave between the spines,
which are stout, about as long as their distance apart
at base and extending upward and backward and
slightly curved inwardly. Petiolar node in profile
cuneiform; above deeply emarginate with the sides
narrowly margined and obtusely angulate. Postpeti-
ole twice as broad as petiole, produced conically at
sides. Gaster short and broad. Legs rather slender.
Opaque. Mandibles punctate, shining. Clypeus
shining, finely striate. Front and cheeks subopaque
with coarse, very rugose striae and the intermedi-
ate spaces rugulose. Scrobes coarsely and granulosely punctate. Vertex
rugosely reticulate. Occipital region with rather regular reticulate striae
and the enclosed portions smooth and shining. Pronotum coarsely, reticu-
lately striate and rugose. Mesonotum more finely rugose, longitudinally striate
at sides. Epinotum with series of fine and short transverse striae. Petiole,
postpetiole, first gastric segment, and posterior portions of second and third
segments very densely punctate and opaque.

Fig. 17. — Pheidole isis
Mann. Soldier.
Front view of head.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

313

Fig. 18. — -Pheidole isis
Mana. Soldier.
Dorsal view of
thorax and petiole.

Sparse erect hairs on head, body, and appendages.

Head and mandibles Uglit brownish red; thorax, epinotum, petiole, and
postpetiole very dark reddish brown, almost black. First gastric segment
brownish yellow, with the posterior border and the remainder of gaster the
same color as thorax. Appendages brownish yellow.

Worker. Length 1.40 mm.

Head a little longer than broad, sides slightly
convex, posterior border feebly concave at middle.
Frontal carinae short. Antennal scapes barely sur-
passing occipital corners; funicular joints two to eight
transverse. Promesonotum rather flat. Pronotal
spines extending outward and forward, a little less
than half as long as their distance apart at base.
Epinotum fiat basally; spines extending upward and
backward and very slightly curved, nearly twice as
long as their distance apart at base. Petiole triangu-
lar in profile, upper border notched at middle. Post-
petiole one and a third times as broad as petiole,
evenly rounded at sides, with straight anterior and
posterior borders.

Opaque. Head, thorax, and abdomen regularly,
densely, and granulosely punctate, the garter more

subtly than the rest; head with sparse striae on front and vertex and
indistinct reticulate striae on occiput.

Hairs sparse, semirecumbent, with a few erect on head and thqrax.

Head light reddish brown. Thorax, petiole, and postpetiole very dark
brown. Gaster and legs brownish yellow.

Female (dealated). Length 3 mm.

Head similar to that of soldier, but broader, with less prominent occipital
corners and less emarginate behind. Pronotum with bluntly conical spines.
Epinotal spines stout, about as long as their distance apart at base. Petiole
shorter than in soldier. Postpetiole broadest in front and subconate.

Head with sculpture similar to that of soldier. Mesothorax irregularly
striate longitudinally and densely punctate. S^utellum more finely striate
and punctate. Petiole, postpetiole, first gastric segment, and posterior
portions of remaining segments densely granulosely punctate.

Pilosity as in soldier.

Color brownish red, with the pronotum, scutellum, a large quadrangular
blotch on mesonotum, thoracic pleurae, petiole, postpetiole, base and apical
border of first gastric segment, and the rest of gaster fuscous. Legs brownish
yellow.

Ysabel: Fulakora.

314

bulletin: museum of comparative zoology.

Near P. tctracantha Emery but is smaller, the clypeus is earinate at
middle, the epinotal spines are shorter and the sculpture and color
entirely different.

The carinae on the head are strongly rugose, the frontal carinae
especially so, having a distinctly crenulate aspect. The elongate
and strongly lobed head and the sculpture are very different from
sexspinosa, though both species belong to the same group.

39a. Pheidole isis Mann, var. taki, var. nov.

Soldier. Length 3 mm. *

Differing from the preceding form only in color, being uniformly pale fer-
ruginous throughout.

San Cristoval: Wai-ai.

40. Pheidole nindi, sp. nov.

Soldier. Length 2.50 mm.

Head a little longer than broad, slightly impressed transversely at vertex,
sides feebly convex, occipital border narrowly excised. Mandibles bidentate
apically. Clypeus nearly fiat, slightly concave at middle of anterior border,

Antennal scapes extending a little more than half
the distance to occipital corners; funicular joints
two to eight broader than long, club elongate, as
long as the remainder of funiculus, with the terminal
joint longer than the two preceding joints together.
Eyes small, moderately convex, situated at anterior
third of sides of head. Promesonotum convex in
profile, sides at middle produced as blunt cones.
Base and declivity of epinotum not separated ; sides
of base with a fine, irregular margin; spines shorter
than their distance apart at base, straight, elongate,
triangular. Petiole thick, node in profile triangular,
narrowly rounded above, margined at sides, dorsal
surface seen from the front very shallowly concave.
Postpetiole transverse, broader than petiole, broadest in front of middle,
with nearly straight sides.

Feebly shining. Mandibles with sparse, regular, and coarse punctures.
Clypeus with a sharp carina at middle and smaller carinae at sides. Front

Fig. 19. — Pheidole nindi
Mann. Soldier. Front
view of head.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 315

and cheeks strongly striate longitudinally. Vertex and occiput coarsely
reticulate, with the interspaces densely punctate. A flat region on sides of
head, inward and posterior to the eyes, cribrately punctate and with a few fine
striae. Promesothorax irregularly reticulate striate and punctate. First
gastric segment finely and very densely striate longitudinally.

Pubescence moderately abundant, long and fine on body, shorter on ap-
pendages.

Brownish red; first gastric segment and appendages yellow-brown.

Worker. Length 1.50 mm.

Head a little longer than broad, as broad in front as behind, sides feebly
convex, posterior border shallowly concave at middle. Clypeus with distinct
anterior and posterior surfaces which are nearly flat; border straight. An-
tennal Sv^apes surpassin,g occipital corners by about twice
their width at tip. First funicular joint longer than the
following three together; joints two to eight about as long
as broad, club slender, as long as remainder of funiculus.
Eyes situated at sides of head a little in front of middle.
Promesothorax little convex above, gibbous at sides, in
profile strongly sloping behind to the mesoepinotal impres-
sion. Base of epinotum longer than the declivity, flat
above; spines rather slender, a little shorter than their
distance apart at base, extending upward, backward, and
slightly outward. Petiole similarly shaped, to that of
soldier but thicker and broader above. Postpetiole Fig. 20.— Pheidole
transverse, twice as broad as petiole, broadest in front J?'" .' t,.^°""

Soldier. Ihorax
of middle. auj^ petiole from

Nearly opaque. Mandibles finely striate and punctate. above.
Clypeus punctate and finely striate. Head, thorax, and
epinotum cribrately punctate, head with fine and sparse striolae, longitudinal
on front and cheeks, somewhat reticulate on vertex and occiput. Petiole,
postpetiole, and first gastric segment shallowly and rather densely punctate.

Pilosity and color as in soldier.

Female. Length 4.75 mm.

fe'

Head broader than long, somewhat narrowed in front, broadly concave
behind.

Head strongly, longitudinally striate, reticulate on occiput, and rather
densely punctate. Mesothorax with finer, dense, and twisted striae and
punctate. Scutellum shining basally with sparse, coarse punctures, rugulose-
striate apically. Epinotum shallowly cribrate-punctate. Petiole cribrately
punctate, the node transversely striate. Postpetiole strongly and closely
striate longitudinally. First gastric segment very densely striolate longi-
tudinally.

316 bulletin: museum of comparative zoology.

Dark brown to black; front of head and anterior third of first gasteric seg-
ment brownish red, tips of femora, tibiae, and tarsi yellowish brown.

San Cristoval: Wai-ai (Type-locality), Pamua, Wainoni Bay. Ugi:
Pawa.

An abundant species in the above localities, where it nests beneath
stones and logs. Tijpe.— M. C. Z. 9^169.

41. Pheidole fuscula Emery.
Term, fuzet., 1900, 23, p. 324, % ^ .

Ysabel: Fulakora.

Several colonies were found in rotten logs.

42. Pheidole philemon Forel.
Rev. Suisse zooL, 1910, 18, p. 44, Q[ ^ c?.

Santa Cruz: Graciosa Bay. Three Sisters: Malapaina. San
Cristoval: Wai-ai. Malaita: Auki.

43. Pheidole umbonata Mayr.

Verh. K. K. zool.-bot. gesellsch. Wien, 1870, 20, p. 977, % .
Ugi: Pawa. San Cristoval: Wai-ai. Malaita: Auki.

44. Pheidole oceanica Mayr.

Sitzungsb. Akad. wiss. Wien, 1866, 53, p. 510, % ^ .
Verh. K.K. zool.-bot. gesellsch. Wien, 1870, 20, p. 979, % ^ .

Ugi: Pawa. Three Sisters: Malapaina. San Cristoval: Wai-ai,
Pamua, Wainoni Bay. Malaita: Auki. Ysabel: Fulakora.

This species, which occurs continuously in all the islands from the
Tonga group to New Guinea, is also abundant in the Solomons. It
lives in large colonies beneath stones or logs or in rotteli wood. My

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 317

specimens agree closely in form and sculpture with a cotype in the
U. S. N. M. collection. In the large series before me there is a great
deal of variation in color, from dark to light brown.

One soldier from Pamua is interesting in having a large and well-
developed median ocellus^ situated slightly to the right of the middle
of the front.

44a. Pheidole oceanica Mayr, var. pattesoni, var. nov.

Santa Cruz: Graciosa Bay.

All of the soldiers and workers in the series from the above locality
are much more uniform and darker in color (dark brown to black)
than specimens from the Solomons and may be considered a geographi-
cal variety. Type.— M. C. Z. 9,162.

45. Cardiocondyla nivalis, sp. nov.

Worker. Length 1.75 mm.

Head longer than broad, narrowest in front; sides moderately convex;
posterior border very shallowly concave. Clypeus flat above and very broadly
and feebly concave at anterior border. Mandibles with three small but dis-
tinct teeth and a larger subapical tooth. Antennal scapes extending four
fifths the distance to occipital corners; first funicular joint as long as the three
following joints together; remaining joints, except those forming the club,
transverse; first joint of club as long as broad; second longer than broad,
terminal as long as the five preceding joints together. Eyes large and convex,
located at sides of head in front of middle. Promesothorax flattened above,
the suture very feeble. Sides of prothorax strongly convex; anterior border
and humeral angles broadly rounded. Mesothorax about as broad as long,
with nearly straight sides. Basal portion of epinotimi rounded above; in
profile straight, and longer than the concave declivity; spines strong, long,
and diverging. Petiole strongly pedunculate, with an acute anteroventral
tooth; node from above rounded in front and at sides; in profile deeper than
long, with declivous anterior surface, nearly flat top and sloping posterior
surface. Postpetiole broader than long; broadest in front of middle, with
convex sides and straight anterior and posterior borders; in profile longer
than high and convex above. Gaster only slightly longer than petiole and
postpetiole together.

Gaster smooth and shining; the rest subopaque and very densely punctulate.
Mandibles subopaque and more coarsely punctate.

318 bulletin: museum of comparative zoology.

Anterior border of clypeus with fine long hairs; funiculus and ventral sur-
face of gaster with sparse and fine pubescence, the rest without pile or pubes-
cence.

Gaster jet black; the rest snowy white or faintly tinged with yellowish brown .

San Cristoval: Pamua.

A striking though minute species, very different from any of the
species known to me. The color of gaster is in strong contrast to the
rest of the body and the long epinotal spines are very distinctive.

Crematogaster subgenus Rhachiocrema, subgen. nov.

In Crematogaster paradoxa Emery from New Guinea and the fol-
lowing new species, the enormous development of the epinotal spines
and the elongate pedunculate structure of the petiole and the elongate
12-jointed antennae with the 2-jointed funicular club, are so different
from other species in the genus that I separate them from Cremato-
gaster sens, strict, as a new subgenus.

The type of the subgenus is : —

46. Crematogaster (Rhachiocrema) wheeleri, sp. nov.

Worker. Length 2.50 mm.

Head with moderately convex sides, broadly rounded occipital corners and
nearly straight border. Mandibular blades with three stout triangular teeth
on apical half. Clypeus convex; anterior border straight. Antennae slender,

Fig. 21. — Crematogaster (Rhaciocrema) wheeleri Mann. Worker. Lateral view of
head, thorax, and petiole.

their scapes surpassing occipital corners; first funicular joint as long as the
succeeding two together; joints two to eight distinctly longer than broad;
club 2-jointed, slender, terminal joint twice as long as the penultimate. Eyes

mann: ants of the British solomon islands. 319

convex, situated at sides of hiead, well behind middle. Promesonotum flat-
tened, without suture; hjameral angles narrowly rounded. Epinotum bearing
two extraordinarily long and massive spines, which are thickened at middle
half, acute apically and divergent, the distance between the tips about equal
to the length of one spine. Petiole long and depressed, thfe node longer than
broad, broadest behind, with posterior corners elevated and obtusely angulate.
Postpetiole broader than long, subangulate at sides.

Shining. Mandibles striate. Clypeus with five or six longitudinal costae.
Remaining parts of head and body finely punctate. Pilosity very long and
sparse; finer, shorter, and semirecumbent on antermae and legs'.

Color black, except mandibles, funiculus, and tarsi which are yellow. The
basal half of funiculus is somewhat fuscous and terminal half clear yellow.

Malaita: near Fourafi in the interior.

The few workers on which the description is based were taken run-
ning about on stones at the edge of a stream where we stopped for
lunch on our walk across the island. Type. — M. C. Z. 9,163.

This species is distinct from paradoxa Emery in the shape of the
spines, which are more thickened, straighter, and less divergent.

47. Crematogaster elysii, sp. nov.

Worker. Length 2 mm.

Head subquadrate, about as long as broad; sides convex, occipital angles
broadly rounded, posterior border nearly straight. Mandibles short, 4-den-
tate. Clypeus convex, feebly concave at middle of anterior border. Antennae

Fig. 22. — Crematogaster elysii Mann. Worker. Lateral view of thorax and petiole.

11-jointed, scapes not attaining occipital border; funicular joints two to seven
moderately transverse, joint eight slightly longer than broad, terminal joint
stout, as long as the three preceding joints together. Pronotum bluntly
margined at sides; humeri angulate. Mesonotum rather flatter than pro-
notum and in a different plane, but not separated by impression. Basal
portion of epinotum broad and shallowly concave, not distinctly separated
from declivity. Epinotal spines half as long as the distance between them

320 bulletin: museum of comparative zoology.

at base, stout and acuminate, curving upward and backward with the tips
inward. Petiolar node a httle longer than broad; broadest in front, margined
at sides and in front; anterior corners rounded; posterior corners angulate.
Postpetiole about as broad as petiole; transverse, longitudinally impressed
above. Gaster short. Legs rather stout.

Smooth and shining throughout. Mandibles punctate; cheeks finely
striolate. Erect hairs absent, except two to three short and stiff hairs on
petiolar node. Head and gaster with very sparse, short, recumbent pubescence.
Funiculus and tarsi pubescent.

Color brownish red; gaster jet black.

Three Sisters: Malapaina.

Described from a few workers taken on recently felled trees. Type.
— M. C. Z. 9,164.

48. Crematogaster abrupta, sp. nov.

Worker. Length 2.50-2.75 mm.

Head slightly longer than broad, as broad in front as behind, with convex
sides, broadly round occipital corners; occipital border very feebly concave
at middle. Mandibles 4-dentate. Clypeus convex; anterior border very
shallowly concave at middle. Antennal scapes not attaining occipital corners;
funicular joints two to seven broader than long, joint eight longer than broad;
terminal joint as long as the three preceding joints together. Eyes large,
little convex, located at sides of head slightly behind middle. Pronotum
flattened medially; sides submargined, humeri rounded. Promesonotal
impression discernible but very feeble. Mesonotum flat basally, then ab-
ruptly declivous, in profile angulate; margined at sides. Epinotum, except
for a narrow flat surface at base, concave, without distinct basal and declivous
portions; spines half as long as their distance apart at base, strong, acuminate,
divergent, curved downward. Petiolar node flat above, longer than broad;
anterior corners rounded, posterior obtusely angulate. Postpetiole transverse,
longitudinally impressed at middle.

Shining. Mandibles subtly striolate. Head, pronotal disc, concave por-
tion of epinotum, and gaster finely and sparsely punctate; front of pronotum
and epinotum, all of mesonotum, petiole, postpetiole, and apical segment of
gaster microscopically rugulose.

Without erect hairs above; with rather sparse and fine recumbent pubescence
on head and gaster; semierect pubescence on scapes and funiculus.

Color uniformly brownish yellow.

Three Sisters: Malapaina.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 321

This form was much more abundant than the preceding and was
found in the same locality. I failed to locate the nests of either.
Tyve.— M.. C. Z. 9,165.

49. Crematogaster foxi, sp. nov.

Worker. Length 2.50-2.75 mm.

Head with strongly convex sides, rounded occipital corners and concave
border. Mandibles 4-dentate. Anterior border of clypeus straight. Anten-
nal scapes extending four fifths the distance to occipital corners; first funicular
joint as long as the three succeeding joints; joints two to eight almost as long
as broad; terminal joint shorter than the three preceding joints. Eyes large
and moderately convex, situated at sides of head posterior to middle. Thorax
slender. Pronotum margined at sides, humeri rounded, disc slightly convex.

Fig. 23. — Crematogaster foxi Mann. Worker. Lateral view of thorax and petiole.

Mesonotum at sides with a narrow elevated margin, the surface shallow, con-
cave, and, posterior to middle, sloping to the deep mesoepinotal impression.
Epinotum shallowly concave; spines subtriangular, acute, half as long as
their distance apart at base and almost straight. Petiolar node slightly longer
than broad, narrowed in front, with roimded anterior and angulate posterior
corners. Postpetiole a little broader than long, longitudinally impressed
above. Gaster elongate.

Shining. Mandibles and cheeks striate. Clypeus with one strong and
several very feeble costae at sides. Pronotum with longitudinal costae, two
of which extend as borders to the mesothorax. Epinotum with indistinct
costae. Petiolar node and base of postpetiole microscopically rugulose, the
rest smooth.

Scattered, very long, and erect yellow pile present everywhere, except on
legs and antennae where it is fine and semirecumbent.

Color yellow.

Female. Length 6.50 mm.

The striation of the mandibles is much coarser than in the worker. The
head and body are smooth and shining, not costate. The wings are hyaline,

322 bulletin: museum of comparative zoology.

with fuscous veins and stigma. The color is ferruginous with a spot on vertex
and the posterior portions of gastric segments fuscous.

San Cristoval: Pamua, Wainoni Bay.

This species is dedicated to Rev. C. R. Fox, missionary to and stu-
dent of the natives on San Cristoval. Type. — M. C. Z. 9,166.

50. Crematogaster nesiotis, sp. nov.

Worker. Length 2.50 mm.

Head with convex sides and shallowly concave occipital border; in the
larger workers broader than long. Mandibles 4-dentate. Clypeus convex,
very shallowly concave at middle of anterior border. Antennae 11-jointed;
scapes not attaining occipital border; funicular joints all longer than broad;

Fig. 24. — Crematogaster nesiotis Mann. Worker. Lateral view of thorax and petiole.

club distinctly .3-jointed, terminal joint about as long as the two preceding
joints together. Thorax robust; promesonotal suture more fully impressed;
pronotum very indefinitely margined at sides, humeri rounded. Mesonotum
in profile convex ; submargined at sides. Flat basal portion of epinotum rather
long, only twice as broad as long; the rest shallowly concave; spines stout and
acuminate, little divergent, curved backward and slightly downward. Petiolar
node as broad or slightly broader than long, concave above, broadest and
narrowly rounded in front of middle. Postpetiole a little broader than long,
the medial furrow narrow.

Shining. Mandibles coarsely striate.

Head and gaster with very sparse recumbent pubescence. Pilosity scattered
on head and body, erect and moderately abundant; shorter and semierect on
legs and antennae.

Color brownish yellow.

Russell: West Bay.

Described from several workers found on tree-trunks. Type. —
M. C. Z. 9,167.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 323

51. Crematogaster obnigra, sp. nov.

Worker. Length 2.25-3 mm.

Head subquadrate, longer than broad, shallowly convex behind, with
rounded occipital corners. Mandibles 4-dentate. Anterior border of clypeus
concave. Antennal scapes extending to half the distance between posterior
border of eyes and occipital corners; funicular joints two to seven about as
broad as long; club slender, terminal joint longer than the two preceding
joints. Eyes little convex, situated at sides of head slightly posterior to middle.
Thorax robust. Pronotum with rounded sides and humeri. Promesonotal
impression indistinct. Mesonotum in profile evenly and rather feebly convex^
sides rounded. Base of epinotum massive, convex, distinct from declivity,
which is one and one half times as long as base and almost flat; spines stout,
triangular, broader at base than long. Petiolar node slightly longer th|an
broad, broadest in front of middle, with narrowly rounded and elevated sides.
Postpetiole broader than long, broadly impressed at middle.

Shining. Mandibles coarsely striate. Chjeeks, anterior half of pronotum,
sides of mesonotum, base of epinotum and meso- and metapleurae striolate.
Petiole and postpetiole punctate, tjie former densely so.

Pilosity moderately abundant, and suberect on head, thorax, abdomen,
legs, and antennae.

Color dark brown to black; mandibles and tarsi fuscous.

Russell: West Bay.
Type.— M. C. Z. 9,168.

Key.

Epinotal spines very short; thorax rounded at sides, not margined; base
of epinotum separated from , declivity by broadly rounded margin;
color black obnigra Mann.

Epinotal spines at least half as long as the distance between their bases;
thorax, at least in part, margined or submargined; base and declivity
of epinotum, not separated 1.

1. Smaller species (length 2 mm.); humeri angulate; without pilosity; bi-

colored, gaster jet black, the rest brownish red elysii Mann.

Length 2.25-3 mm. humeri rounded, not bicolored 2.

2. Petiolar node broadest behind; clypeus bicostate; pronotum and meso-

notum sharply costate; gaster unusually long and slender, .foxi Mann.

Petiolar node broadest in front; thorax not costate; gaster of ordinary

form 3 .

324 bulletin: museum of comparative zoology.

3. Mesothorax with sides margined, in profile angulate 'behind middle, the
posterior face almost perpendicular to the mesoepinotal impression;

body without erect hairs above abrupta Mann.

Mesonotum not margined, evenly convex in profile; body pilose above.

nesiotis Mann.

52. VoLLENHOViA PEDESTRis (F. Smith).

Myrmica pedestris F. Smith, Journ. proc. Linn. soc. London. Zool., 1862, 6,
p. 46, y .

Santa Cruz : Graciosa Bay. Santa Anna. San Cristoval : Wai-ai
Pamua, Wainoni Bay. Ugi: Pawa. Malaita: Auki. Ysabel:
Fulakora.

This is the most widely distributed species of Vollenhovia and one
of the commonest ants in the Solomons. The colonies are found
beneath bark and in rotten wood where some humidity is present.
They make well-defined runways beneath the bark and move along
these, more or less in file. One colony, seen at Ugi, was composed of
many thousands of individuals and occupied the entire interior of a
hollowed log lying on the ground. In collecting from this nest I was
stung repeatedly, but the sting was not severe.

53. Vollenhovia subtilis Emery.

Ann. Mus. civ. stor. nat. Genova, 1887, 25, p. 454, y 9 .

San Cristoval: Pamua, Wai-ai, Wainoni Bay. Malaita: Auki.

Found nesting beneath bark. My specimens agree with Emery's
description of workers of the typical form from Key Island which have
the epinotimi smooth and shining at middle.

The variety affinis Emery described from New Guinea is not repre-
sented among my material.

54. Vollenhovia loboii, sp. nov.
Worker. Length L50 mm.

Head much longer than broad, with subparallel sides and rather narrowly
concave occipital border. Mandibles 5-dentate. Antennae short, their

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 325

scapes extending about two thirds the distance to occipital corners; first
funicular joint as long as the three succeeding joints together; joints two to
seven strongly transverse; club large, longer than the rest of funiculus. Eyes
little convex, situated at sides of head well in front of middle. Promcsonotum
flattened; humeri bluntly angulate. Mesoepinotal suture moderately im-
pressed. Basal portion of epinotum fiat, longer than the declivity; declivity
flat, with blunt, elevated margins laterally. Petiolar node a little longer than
broad; in profile slightly longer than high, with sloping anterior and rounded
posterior surfaces. Postpetiole subglobose, a little broader than petiole.

Feebly shining throughout. Mandibles sparsely punctate. Head reticu-
lately striate and punctate; clypeus smooth. Promesonotum and epinotum
striolate and punctate, more finely and shallowly than the head, with the base
of epinotum as strongly so as the rest; thoracic pleurae cribrately punctate;
petiole and postpetiole densely but more shallowly punctate. Gaster with
fine piligerous punctures.

Fine, long, erect yellowish hairs moderately abundant on head and body;
shorter and semirecumbent on appendages.

Dark brown to black; antennae and legs yellowish brown; femora inf uscated
at basal two thirds.

Malaita: Auki.

Near brevicornis Emery but with the scapes a little longer and the
sculpture entirely different. The small size, more elongate head a;nd
the sculpture of the basal portion of the epinotum distinguish it
from subtilis Emery and its varieties.

55. VOLLENHOVIA DENTATA, Sp. nOV.

Worker. Length 1.75-2 mm.

Head about a fourth longer than broad, sides subparallel, occipital corners
broadly rounded, border shallowly concave at middle. Mandibles 5-dentate.
Clypeus convex, rounded above and at anterior border. Antennal scapes
extending three fourths the distance to occipital corners; first funicular joint
as long as the three following joints together; joints two to seven moderately
transverse; club large. Eyes slightly convex, situated at sides of head a little
in front of middle. Promesonotum flattened; humeri obtusely angulate.
Mesoepinotal suture moderately impressed. Basal and declivous surface of
epinotum rounding into each other; sides of declivity with margin that is
elevated at middle into a small, stout, triangular tooth. Petiolar node a
little longer than broad ; in profile about as high as long, anterior face declivous,
dorsal and posterior surfaces moderately rounded; antero ventral tooth broad

326 bulletin: museum of compakative zoology.

and thin, rounded at tip. Postpetiole one and a third times as broad as
petiole and a Uttle broader than long; in profile, as long as high, rounded
above. Legs less swollen than in the related species.

Feebly shining. Mandibles with sparse, fine punctures. Head, thorax,
and base of epinotum rugosely punctate and reticulately striate. Petiole,
postpetiole, and first gastric segment with foveolate punctures which are more
regular and shallower on the gaster.

Fine, erect, yellowish hairs moderately abundant on head and body, and
shorter, stiffer, semierect ones on appendages.

Dark, reddish brown, gaster black; mandibles and appendages yellowish
brown.

Male. Length 2.25 mm.

Head, excluding eyes a little longer than broad, corners broadly rounded,
occipital border nearly straight. Mandibles elongate, flat and broadly rounded
at tips. Clypeus strongly convex, narrowly rounded in front. Eyes large
and convex, more than half as long as head; situated at a distance equal to
one fifth their length from the anterior borders of head. Ocelli large. An-
tennae slender, 13-jointed, their scapes two thirds as long as the eyes; first
funicular joint scarcely longer than the second; all joints longer than broad,
increasing in length toward apex; apical joint slender, about as long as the
two preceding joints together. Thorax robust. Mesothorax flat. Mayrian
furrows not present; parapsidal furrows short and broad. Scutellum trans-
verse, flattened. Epinotum with a rounded angle between base and declivity.
Petiole in profile longer than high, broadly rounded above; beneath with a
minute sharp tooth in front of middle; from above, longer than broad,
with moderately rounded sides. Postpetiole subglobose, a little broader
than petiole. Gaster narrow. Genitalia prominent, legs long and very
slender.

Feebly shining. Head subopaque, densely punctate. Thorax, petiole,
postpetiole, and first gastric segment finely, shallowly, and rather densely
punctate, the gaster less densely than the rest.

Short, silky hairs abundant on head, body, and appendages.

Color dark brown to black; legs brown, antennae yellowish brown. Wings
evenly infuscated.

Ugi: Pawa.

This species differs from the other Papuasian forms in the denticu-
late margins of the epinotal dechvity, in the coarse punctation of the
first gastric segment, and the generally coarser sculpture of head and
thorax. It was fairly common on Ugi, nesting beneath bark like the
other species. I did not find it on other islands, but took the following
subspecies on Malaita. Typc.— M. C. Z. 9,170.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 327

55a. VoLLENHOviA DENTATA Mann, subsp. MARGINATA, subsp. nov.

JVorkcr. Length 2 mm.

Differing from dentata in having the teeth at margins of epinotal decUvity
longer and narrower. The anterior upper border of the petiohir node is dis-
tinctly margined. The sculpture of the promesothorax is feebler, especially
in the middle where it is reduced to delicate strialation. The petiole and post-
petiole are irregularly and finely strialate and the gaster is not punctate.

The whole body is more shining. The color is light brownish red with the
appendages honey-yellow.

Malaita: Interior.

Described from a single worker taken at our camp at 2,300 feet
elevation on the trail between Atta and Fourafi.

56. VoLLENHOVIA ELYSII, Sp. nOV.

Worker. Length L6 mm.

Near brevicornis Emery. Head about one and a fourth times as long as
broad, slightly convex at sides and concave at posterior border. Mandibles
5-dentate. Antennae short, their scapes extending a little less than two thirds
the distance to occipital corners. Promesothorax slender and flattened;
humeri very obtusely angulate. Mesoepinotal suture but feebly impressed.
Epinotal declivity at sides with a coarse margin that is elevated at middle
into a broad, low triangle. Petiolar node a little longer than broad; in profile
straight in front, rounded above anteriorly and sloping behind; ventral tooth
twice as broad as long and only slightly rounded at tip. Postpetiole sub-
globose, a little broader than petiole.

Moderately shining. Mandibles sparsely punctate. Cljrpeus with delicate,
interrupted striae. Head, thorax, and base of epinotum irregularly strialate
and punctate. Petiole indistinctly striate transversely. Postpetiole finely
and shallowly punctate. Gaster with sparse piligerous punctures.

Hairs on head and body abundant, long, and fine; on appendages shorter
and semirecumbent.

Color reddish brown ; dark on head and light on gaster.

Three Sisters: Malapaina.

In its short antennal scapes elysii resembles brevicornis Emery, but
differs in the much smaller size, in having the sculpture of the thorax
uniform, and in the elevated angular margins to epinotal declivity.

328 bulletin: museum of comparative zoology.

The sculpture is more delicate than in the other species I have seen.
The diflFerence in sculpture among the species of Vollenhovia is easy to
see but difficult to describe.

57. Vollenhovia foveaceps, sp. nov.
Worker. Length 1.50 mm.

Head longer than broad, slightly narrowed in front, with feebly convex sides
and rather strongly concave posterior border. Mandibles 5-dentate. Clypeus
nearly twice as long as broad, with flattened surface. Antennal scapes thick-
ened distally, extending about three fourths the distance to occipital corners;
funicular joints, except the first and those forming the club, strongly trans-
verse. Eyes convex, situated well in front of middle of sides of head. Thorax
flattened, with prominently angulate humeri. Mesoepinotal suture feebly
discernible from above. Basal and declivous portions of epinotum rounding
into each other, the surface of the latter flat. Petiolar node from above a
little longer than broad, with straight sides; in profile, thick, about as high as
long, highest at anterior margin, broadly rounded above and behind and
declivous in front, anteroventrally with a stout tooth which is broad and
rounded apically. Postpetiole nearly twice as broad as petiole, a little broader
than long and evenly rounded at sides; in profile as high as long and rounded
above.

Gaster elongate oval. Legs long, femora and tibiae thick.

Shining throughout. Mandibles impunctate. Head with foveolate punc-
tures which are sparse on occiput and vertex and more abundant in front;
front and sides delicately, reticulately striolate, except for a narrow median
smooth space extending down front and vertex. Pronotum, mesonotum,
sides of epinotum, petiole, and postpetiole with very sparse but strong foveae.
The rest smooth, though under a strong lens a subtle reticulation is discernible.

Scattered, sparse, very long, fine, and erect hairs present on head, body,
and antennae, and shorter semirecumbent hairs on legs.

Head light brown with a darker blotch on vertex; legs and antennae yellow-
ish, the rest dark brownish red. Pilosity yellowish.

/^ewa/e (dealated). Length 1.8 mm.

Very similar to the worker, with the usual sexual modifications. The ocelli
as large as the cephalic foveae. Pronotum with angulate humeri. Mesono-
tum flat, transversely oval. «

Thorax, epinotum, petiole, and postpetiole with scattered foveolate punc-
tures as in worker, mesonotum with more abundant punctures.

Pilosity more abundant than in worker. Color darker.

Ysabel: Fulakora.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 329

Described from females and workers from two small colonies that
contained less than a dozen workers each.

In the very feebly impressed mesoepinotal suture, the sharply
angulate pronotal humeri and in sculpture, foveaceps and the closely
related punctata Vieh. from New Guinea differ markedly from the
other Papuasian species. V. punctata is larger than foveaceps and
differently colored and has the petiole and postpetiole similar to
subtilis Emery subsp. magna Viehmeyer. In that form, accoi'ding to
Viehmeyer's description, the postpetiole is somewhat broader than
the petiole and has strongly convex sides. In foveaceps it is twice as
broad as the petiole and the sides are nearly straight.

58. MoNOMORiUM TALPA Emery.

Lorentz's Nova Guinea, 1911, 9, p. 2,52, ^ .

Female. Length 3 mm.

Head rugulosely striate and punctate. Thorax and node coarsely punctate.
Epinotum in profile angulate similarly to worker. Petiole and postpetiole
thicker than in worker.

Color dark reddish brown; mandibles, antennae, and legs brownish yellow.

San Cristoval: Star Harbor, Wainoni Bay. Ugi: Pawa.
Occurs beneath stones and in rotten logs.

59. MoNOMORiUM PHARAONis (Linue).
Formica pharaonis Linne, Syst. nat., ed. 10, 1758, 1, p. 580.
Malaita: Auki. Ugi: Pawa.^

60. SOLENOPSIS PAWAENSIS, Sp. nOV.

Worker. Length 1 mm.

Head a little longer than broad, broadest behind, with moderately convex
sides and concave occipital border. Clypeus bicarinate and bidentate, the
teeth broad and triangular. Mandibles with four large teeth. Antennal
scapes extending three fourths the distance to occipital corners; funicular

330 bulletin: museum of comparative zoology.

joints two to six moderately transverse; club slender, as long as the rest of
funiculus, terminal joint nearly four times as long as the penultimate. Eyes
minute, composed of only one facet. Promesonotum a little longer than broad,
rounded above and at sides. Mesoepinotal impression profound. Epinotum
with equal base and declivity, the former flattened, the latter slightly concave
and margined laterally. Petiolar node in profile subtriangular, rounded above;
about as long as the pedicel; from above broader than long. Postpetiole
slightly broader than petiole, and one and one third times as broad as long.

Shining throughout. Epinotum and pedicel of petiole coarsely and densely,
though shallowly, punctate and less shining than the other parts.

Front and clyjDeus with a few long and coarse hairs. Head and gaster
sparsely and microscopically pubescent.

Color yellow to yellow-brown.

Ugi: Pawa (Type-locality). Ysabel: Fulakora.

The Pawa specimens are darker than those from Fulakora but other-
wise identical.

Solcnojms ixncaensis resembles yaimana Emery, but differs in its
minute eyes, the narrower petiole and in the coarse punctuation of the
epinotum and pedicel.

61. SOLENOPSIS CLEPTIS, Sp. nOV.

Worker. Length 1 mm.

Head quadrangular, longer than broad, as broad in front as behind, with
nearly straight sides, broadly rounded occipital corners and straight border.
Mandibles stout, 3-dentate. Clypeus bicarinate, the carinae Uttle divergent,
terminating in acuminate teeth. Antennae short, scapes extending less than
two thirds the distance to occipital corners; funicular joints two to six twice
as broad as long; terminal joint slender, as long as the rest of funiculus. Eyes
situated at front fourth of head, as long as the diameter of scape at base. Pro-
mesothorax one and a half times as long as broad; pronotum with rounded
sides. Basal portion of epinotum convex and rounding into the declivity
which is convex and three times as long as the base., Petiolar node in profile
with nearly straight anterior surface; the top and posterior surface round into
each other; ventral surface strongly bisinuate; peduncle much shorter than
the node; node from above subglobose, as broad as long. Postpetiole a
little broader than the node and a little broader than long.

Shining, with sparse, shallow punctures, each with a long, erect hair. Legs
and antennae with shorter, suberect hairs. Mandibles sparsely punctate.

Color brownish red; appendages lighter and gaster somewhat darker.

San Cristoval : Wai-ai.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 331

Solenopsu dahli Forel and maxillosa Emery are similar to cleptis, but
the latter is distinct in having the antennal scapes much shorter and
in the bisinuate ventral surface of the petiole.

62. Oligomyrmex atomus Emery.
Term, fuzet., 1900, 23, p. 328, pi. 7, fig. 30, %.

Female (dealated). Length 2.75 mm.

Form long and slender. Head about a fourth longer than broad, occipital
border less concave than in worker and unarmed. Ocelli large. Eyes large
and but little convex, situated at sides of head at a distance of less than half
their length from anterior margins. Thorax flat above, slender. Epinotiun
without distinct base and declivity, flattened, unarmed. Petiole and post-
petiole as in soldier. Gaster subcyhndrical, three times as long as broad.

Sculpture of head as in worker. Thorax and epinotum shining, sparsely
though rather coarsely punctate. Petiole and postpetiole rather densely
punctate. Gaster finely and regularly punctate.

Color reddish brown; appendages lighter.

Ysabel: Fulakora.

Several colonies of this minute species were found in rotten wood.
The soldiers agree closely with Emery's description but the occipital
border is rather more deeply concave and the mandibular teeth larger
and more regular than shown in Emery's figure. The color is darker
and comparison with specimens from New Guinea may show the
Solomon Island specimens to be different.

63. Oligomyrmex viehmeyeri, sp. nov.

Soldier. Length, 2.50 mm.

Head about a third longer than broad, with subparallel sides and broadly
rounded occipital corners; occipital border strongly concave at middle, upper
borders of concavity armed with short denticulate processes. Mandibles
with four stout teeth and two smaller ones. Basal portion of clypeus flat and
separated by an angle from the anterior part which is slightly concave, anterior
border broadly rounded. Frontal area large, triangular. Antennal scapes
flattened basally, extending less than half the distance to occipital corners;
funicular joints three to seven a little broader than long, joints eight and nine
about as long as broad; club a little shorter than remainder of funiculus, termi-

332

bulletin: museum of comparative zoology.

nal joint nearly three times as long as penultimate. Eyes oval, twice as long as-
broad, very flat, situated at anterior third of sides of head. Pronotum elevated,
strongly convex above and at sides. Promesonotal suture feebly impressed.
Mesonotum sUghtly convex. Mesoepinotal suture profoundly impressed. Epi-
notimi concave between the spines, which are
triangular and acute apically, about twice as long
as their width at base and extending upward and
slightly forward. Petiolar node squamiform, emar-
ginate at middle of upper border; from above three
times as broad as long. Postpetiole a little broader
than petiole, broadest in front and narrowly rounded
at sides, gaster elongate oval.

Shining. Mandibles sparsely punctate. Middle of
clypeus and frontal area smooth. Sides of clypeus,
cheeks, and front longitudinally striate and finely
rugulose between the striae; vertex and occiput
rugulose, with indistinct striae. Promesonotum
rugulose, pleurae, epinotum, and peduncle of petiole
cribrately punctate. Petiolar node finely punctate.
Postpetiole and base of first gastric segment rugosely
punctate, remainder of first gastric segment with regular punctures.

Recumbent silky pubescence moderately abundant on head and body and
sparse, coarser, erect pile on front of head, petiole, postpetiole, and apical por-
tion of gaster.

Fuscof erruginous ; mandibles and appendages lighter.

Fig. 25. — Oligomyrmex
viehmeyeri Mann. Sol-
dier. Front view of head.

Worker. Length .75-1 mm.

Head a little longer than broad, with convex sides and nearly straight
occipital border. Mandibles with five rather large triangular teeth. Anten-
nal scapes extending three fourths the distance to occipital corners; middle
funicular joints transverse; club as long as remainder of funiculus, terminal
joint three times as long as penultimate. Eyes small, situated at sides of
head well in front of middle. Promesonotum subglobose, as long as broad.
Epinotum flattened at base, the spines slender, acute, a little less than half as
long as their distance apart at base. Petiolar node higher than long, twice as
broad as long and rounded above. Postpetiole as broad as petiole, transversely
oval. Gaster short and broad. Legs long.

Shining. Cheeks striate. Epinotum cribrately punctate, the rest very
finely punctate.

Fine short, recumbent pilosity sparsely distributed on head, body, and
appendages.

Color light ferruginous; appendages yellowish.

Female. Length 2.80 mm.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 333

Form slender. Head much longer than broad, sides subparallel, posterior
border feebly emarginate. Ocelli prominent. Eyes large and rather flat.
Thorax two and a half times as long as broad, broadest at wing insertions,
flattened above. Petiolar node shorter and thicker than in soldier and not
emarginate above. Postpetiole sUghtly broader than petiole, transversely
oval. Gaster three times as long as broad.

Shining. Head punctate similarly to that of soldier. Epinotum rugulose.
Petiolar node densely and shallowly punctate, the remainder with sparse, fine
punctation.

Pilosity of head as in soldier, of thorax and abdomen more abundant, erect
on thorax and suberect on abdomen.

Color f uscof erruginous ; appendages lighter. Wings hyaline, veins brown.

Male. Length 2.5 mm.

Head, excluding eyes, much longer than broad, concave at occipital border.
Mandibles well developed, 5-dentate. Cl3T3eus flattened at middle. An-
tennae (broken in specimens before me) with scape twice as long as first
funicular joint, joints three to ten cylindrical, each as long as the eye. Eyes
very large and convex. Ocelli large. Thorax twice as long as its width at
wing insertions. Scutellum slightly convex. Petiole about twice as long as
broad, node low and rounded. Postpetiole barely broader than petiole and
as broad as long, narrowed in front. Genitalia prominent.

Shining. Head very finely rugosely striolate, thorax and abdomen very
minutely punctate; postpetiole rather densely so. Hairs on head and thorax
short, stiff, and erect; on gaster silky and recumbent. Funicular joints with
a dense covering of short, white hairs.

Ferruginous; head epinotum, petiole, postpetiole, and first gastric segment
fuscous.

Wings hyaline, with dense fuscous hairs, veins brown.

San Cristoval: Wai-ai.

Described from a series taken from several colonies in rotten wood.
The worker resembles that of 0. subreptor Emery of New Guinea
which is known only from that phase, but has very pronounced epino-
tal spines, while in subreptor the sides are merely obtusely angulate.
Ttjpe.— M.C.Z. 9,171.

64a. PoDOMYRMA BASALis Smith, subsp. SALOMO, subsp. nov.

Worker. Length 5.5-6.5 mm.

In general appearance very similar to the typical Australian form, but differ-
ing in the following characters : —

The striae of the head, instead of being strong and uniform, are very delicate,

334 bulletin: museum of comparative zoology.

and, on the vertex, almost obsolete, and interspersed with coarse, shallow punc-
tures.

The occipital region is strongly infuscated and the femora are bright ferru-
ginous, except at the apical third, where they are black. The apical band on
the first gastric segment is narrow, though with a tendency to extend forward
as an obscure median line.

Florida: Tulagi. Malaita: Auki.

64b. PoDOMYRMA BASALis Smith, subsp. wooDFORDi, subsp. nov.

Worker. Length 3.50-4.75 mm.

Head quadrate, a little longer than broad, as broad in front as behind, sides
nearly straight, occipital border broadly and shallowly concave. Mandibles
with five stout teeth. Clypeus rather flat, anterior border nearly straight.
Frontal carinae very feeble posteriorly, little divergent, extending nearly to
occipital border. Antennal scapes stout, extending a little more than two
thirds the distance to occipital corners; funicular joints two to eight slightly
longer than broad; club shorter than remainder of funiculus, with the termina 1
joint as long as the two preceding joints together. Eyes large, little convex,

Fig. 26. — Podomyrma basalis woodfordi Mann. Worker.

situated at sides of head posterior to middle. Pronotum depressed in front,
humeral angles broader at base than long, triangular, not very acute at tips.
Mesonotal region sloping. Basal portion of epinotum slightly convex, broadly
rounding into the declivity. Petiole from above about three times as long as
broad, subquadrate; in profile convex at middle, armed dorsoanteriorly with
a thick conical spine and anteroventrally with a broader, flat spine, rounded
at tip. Femora strongly incrassate, tibiae moderately so.

Moderately shining. Mandibles coarsely striate. Clypeus with three
distinct and entire striae on either side and much feebler and shorter striae on
middle portion. Striae of head fine, entire, and longitudinal. Pronotum with
a series of nine longitudinal, and sHghtly sinuous costae, six of which extend
on to the mesonotum. Epinotum with a low, rounded ridge on either side.
Petiole with acute lateral carinae. Basal gastric segment with short striae at
middle of base.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 335

Hairs on head, and body and appendages stiff, erect, and sparse, on scapes
with very sparse short and fine recumbent hairs.

Color yellow to yellow-brown; the head and thorax darker and the geni-
culae strongly infuscated.

Female. Length 11 mm.

Head similar to that of worker. Ocelli small. Mesothorax nearly twice as
long as broad, truncate behind and narrowly rounded in front ; in profile con-
vex at anterior half, flat posteriorly. Scutellum transverse, the surface flat-
tened. Base of epinotum strongly convex; declivity flat. Petiole lacking
the anterodorsal projection; in profile two and a half times as long as broad,
highest a little in front of middle, with the anterior surface broadly rounding
into the dorsum. Postpetiole differing from that of the worker in being
broadly instead of narrowly rounded above in profile, with the anterior and
posterior surfaces less flat, and in lacking the median impression on the pos-
terior surface. Gaster elongate.

Very shining. Head and mandibles sculptured similar to but more coarsely
than that of worker. Prothorax superficially and longitudinally costate.
Mesothorax smooth, except apically, where there are a few shallow striae in the
middle. Epinotal base costate at sides, declivity smooth. Sides of petiole
and posterior portion of postpetiole feebly costate, the rest smooth. Gaster
and legs as in worker.

Head with fuscous blotches at ocelli. The fuscous marking of the first
gastric segment extends at middle for half the length of the segment. Color
otherwise as in worker.

Wings (length 8 mm.) almost clear; veins and stigma pale brown.

Three Sisters: Malapaina.

Described from many workers and two females. Type. — M. C. Z.
9,172.

This form is distinct from hasalis in size, sculpture, and color, it was
very abundant on Malapaina, and probably occurs elsewhere in the
group, but was found only in the tops of recently felled trees.

The colonies live in twigs about an inch in diameter, in chambers
evidently made by wood-boring beetles and the same colony may
occupy more than one of these chambers, which have no passages
between.

65. Myrmecina modesta, sp. nov.

Worker. Length 2.50-2.75 mm.

Head a little longer than broad, with slightly convex sides and shallowly
concave occipital border. Clypeus broadly and transversely carinate at

336 bulletin: museum of comparative zoology,

middle; bidentate at anterior border. Mandibles stout, their blades armed
with six or seven small teeth and a large subapical tooth about as long as the
terminal one. Antennal scapes extending to the occipital corners; first joint
of funiculus three times as long as the very short second joint; joints three to
eight transverse, but not strongly so ; club as long as the rest of the funiculus,
the terminal joint longer than the two others together. Prothorax three times
as broad as epinotum; humei'al and inferior corners angulate. Epinotal base
sloping and separated from declivity by a transverse border; declivity convex;
spines stout at base and acuminate at apex, as long as the declivity. Node,
from above, quadrangular, anterior corners angulate; in profile longer than
high, with a sloping anterior face equal in length to the straight dorsal surface.
Postpetiole broader than long and quadrangular.

Somewhat shining. Head with coarse striae, which are straight and parallel
in front and somewhat irregular and oblique at sides. Clypeus and mandibles
finely punctate and shining. Striae of thorax strong, longitudinal, and extend-
ing to epinotum where they become more feeble apically. Epinotal declivity
smooth and shining. Petiole and postpetiole with widely separated costae,
the spaces between smooth and shining. Legs sparsely punctate and shining.
Gaster finely punctate and more shining than the other parts.

Head, body, mandibles, and legs with very abundantly and finely pilose.

Black. Mandibles, antennae, and legs brownish red, in some specimens the
lower part of femora darker than the rest. Pilosity gray.

Mab. Length 2.75 mm.

Head, excluding eyes, as broad as long, and broadly rounded behind.
Cheeks very short. Clypeus transverse, with truncate anterior border.
Antennal scape as long as the first two funicular joints together; first funicular
joint a little more than half as long as second; second joint twice as long as
broad and a little longer than the third; joints three to eleven subequal,
each a little less than twice as long as broad; terminal joint longer than the
two preceding joints together. Prothorax convex in front', flattened behind,
with strong Mayrian furrows. Scutellum transverse and moderately convex.
Epinotum with equal base and declivity, which are separated by an acute
margin; spines stout and triangular. Petiole longer than in the worker, and
the anterior surface proportionally shorter.

Shining. Head smooth except for a few fine striae inward from the eyes
and between the ocelli, and minute punctures. Pro- and mesothorax very
finely punctate and with sparse and coarse punctures apically. Scutellum
almost smooth on disc, rugose laterally. Base of epinotum with six or seven
strong longitudinal striae. Petiole and postpetiole sculptured as in worker.
Gaster very shining.

Head, body, and appendages with abundant, long, fuscous pile.

Black; mandibles, antennae, and legs rufous, femora darker; genitalia white.
Wings pilose, slightly infuscated, with brown veins and stigma.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 337

Santa Cruz: Graciosa Bay (T^^pe-locality). Three Sisters: Mala-
paina. Ugi: Pawa. San Cristoval: Wainoni Bay, Wai-ai. Ysabel:
Fulakora.

Occurs in small colonies beneath stones. 31. mandibularis Vieh-
meyer from New Guinea is very close to modcsta, but has on the inner
edge of the mandibles a broad widening, which Viehmeyer compares
with that of Acropyga hiddli Forel. In moclcda the posterior corners
of the blades are produced and rounded continuously with the inner
border, resembling the structure of the mandible of Acropyga termi-
fobia, rather than of butclli, as figured by Forel. Type. — M. C. Z.
9,186.

65a. Myrmecina modesta Mann, subsp. subarmata, subsp. nov.

Worker. Length 2.50-2.75 mm.

Differing from the preceding in having the epinotal spines much reduced in
size, being less than half as long as the epinotal declivity, triangular in shape
and not acute at apex.

Malaita: Auki.

T^jpe.— M.C.Z. 9,187.

Key to the Papuasian Species.

Body without strong sculpture; epinotal spines long and acute. (New

Guinea) polita Emery.

At least the thorax coarsely sculptured 1.

1. Gaster opaque, subtly rugosely striate longitudinally. Length 3.5 mm.

(New Guinea) ^ opaciventris Emery.

Gaster shining. Length 2-3 mm 2.

2. Antennal scapes not attaining occipital corners; anterior border of clypeus

unarmed. (New Guinea) brevicornis Emery.

Antennal scapes attaining or surpassing occipital corners; clypeus den-
tate 3.

3. Pronotum arcuately transversely striate; clypeus obscurely 4-dentate

(New Guinea) transversa Emery.

Pronotum longitudinally striate or punctate 4.

4. Ferruginous; head and promesonotum with pihgerous punctures; clypeus

3-dentate. (New Guinea) punctata Emery.

Black; head and promesonotum coarsely striate; clypeus 2-dentate. . .5.

338 bulletin: museum of comparative zoology.

5. Inner edge of mandibles with a broad, blunt thickening. (New Guinea).

mandihularis Viehmeyer.

Inner edge of mandibles swollen apically, but the swollen part continuous
with the blade 6.

6. Epinotal spines as long as the declivity and acute (Santa Cruz).

ynodesta Mann.
Epinotal spines very short and not acute. (Solomons).

viodesta subsp. subarmata Mann.

66. Pristomyrmex pegasus, sp. nov.

Worker. Length 1.75 mm.

Head, including mandibles, a little longer than broad, with slightly convex
sides, broadly rounded corners and nearly straight occipital border. Mandi-
bles strongly curved, their basal borders bluntly dentate anterior to middle,
blades acutely 4-dentate. Clypeus concave; with a very short carina basally ;
anterior border bearing a tooth at each side and feebly bisinuate but not
toothed at middle. Antennal scrobes strong, extending two thirds the dis-

FiG. 27. — Pristomyrmex pegasus Mann. Worker. Lateral view of thorax and petiole.

tance to occipital corners. Antennal scapes not attaining occipital corners;
funicular joints two to seven a little broader than long; terminal joint slender,
a little longer than the two preceding taken together. Promesonotum robust,
broadly and transversely impressed in front, longitudinally impressed behind,
both impressions very shallow; sides of mesonotum elevated behind into thin
triangular teeth. Epinotum with equal base and declivity, the former trans-
versely carinate apically; spines strong, as long as declivity. Petiolar node
in profile higher than long, narrowly rounded above, anterior surface concave,
posterior convex; peduncle short, from above nearly twice as long as broad.
Postpetiole rounded, in profile higher than long.

Shining. Mandibles coarsely, sparsely punctate. Head and thorax
foveolately punctate, the punctures being coarser and more abundant on the
head; clypeus subopaque, denselj^ punctulate; scrobes transversely carinate

MANN : ANTS OF THE BRITISH SOLOMON ISLANDS. 339

in front, smooth behind. Rest of body minutely punctate and very shining.
Everywhere with sparse, scattered, rather stiff, suberect hairs. Pile yel-
lowish.

Santa Cruz: Graciosa Bay.

Only one worker was found. The elevated sides of the mesothorax
and the absence of the median tooth on anterior border of clypeus
is characteristic of this distinct species.

67. PRISTOMYRMEX OBESUS, Sp. nOV.

JVorkcr. Length 2 mm.

Head as broad as long, with convex sides, broadly rounded corners and
straight occipital border. Clypeus flat in middle; anterior border with three
triangular teeth. Mandibles stout, their basal border dentate at middle;
blades strongly 3-dentate. Eyes little convex, located at middle of sides of
head. Antennal scrobes broad and shallow, extending half the distance from
eyes to occipital corners. Antennae short; scapes somewhat thickened in
front of middle and at apex, extending five sixths the distance to occipital
corners; first funicular joint much broader and nearly twice as long as the
second; joints two to seven moderately transverse; eight and nine a little
longer than broad; terminal joint two and one half times as long as broad,
broadest in middle and acuminate anieriorly. Promesonotum with a trans-
verse impression in front which is joined at middle by a deeper longitudinal
impression; anterior border thinly margined, sides more roundly margined
and bluntly bituberculate ; inferior corners evenly rounded. Epinotum with
subequal base and declivity, the base transversely carinate at middle; de-
clivity concave, margined at sides; spines triangular, acute and nearly as long
as declivity. Peduncle of petiole nearly as long as the node; node in profile
deeper than long and highest in front, with concave anterior and posterior
surfaces and slightly convex dorsum; from above, twice as long as broad,
sides straight. Postpetiole from above, as broad as long; in profile, deeper
than long and rounded above.

Shining. Mandibles with sparse and moderately strong punctures. Head,
except clypeus and antennal scrobes which are smooth and shining, coarsely,
foveolately punctate. Thorax punctate similarly to head but much more
sparsely. Epinotum, petiole, postpetiole, and gaster very shining, with a few
shallow punctures.

Legs and antennae finely punctate. Every where with scattered, fine, erect
hairs.

Color ferruginous.

Female (dealated). Length 2.5 mm.

340 bulletin: museum of comparative zoology.

Differing from the worker in having the epinotal spines shorter, the petiolar
node thicker and less elevated. The mesothoracic punctures are coarser.

Male. Length 2 mm.

Head, excluding eyes, as long as broad, broadly rounded behind. Clypeus
slightly convex, narrowly rounded in front. Mandibles very feeble. Eyes
less than half as long as head, strongly convex, situated at sides at a distance
of two thirds their length from base of clypeus. Antennae stout; scape shorter
than eyes; first funicular joint half as long as second; joints two to ten sub-
equal, cylindrical, about twice as long as broad; terminal joint as long as the
two preceding joints. Mesonotum with strong Mayrian furrows; longitudi-
nally impressed at middle. Scutellum broader than long, slightly convex.
Epinotum with distinct base and declivity; tuberculate at sides. Petiole
from above more than twice as long as broad; in profile, slender, twice as long
as high, the node evenly rounded and grading into the peduncle. Postpetiole
rounded, as long as deep and a little longer than broad. Legs slender.

Shining, finely punctate, with rather stiff black hairs scattered on head
and body, legs, and scape; funiculus and legs with fine white pubescence.

Color black. Wings strongly infuscated and hairy, veins, stigma, and hairs
fuscous.

Ysabel: Fulakora (T^'pe-locality). Malaita: Auki. Three Sis-
ters : Malapaina.

The only colony that I found, beneath a stone at Fulakora, was a
small one, composed of less than a dozen workers, a dealated female,
and one male. — Type. — M. C. Z. 9,173.

In this small series there is some slight variation in the length of the
epinotal .spines.

Pri^tomyrmex obestis differs from quadridens Emery and coggii
Emery in not having teeth on the sides of pronotum. The epinotal
spines are shorter and thicker in typical ohesus, but in the following
subspecies they are more similar to those of quadridens.

67a. Pristomyrmex obesus Mann, subsp. melanoticus, subsp. nov.

Worker. Length, 2 mm.

Differ from the typical form in the structure of the epinotal spines, which
are much more slender and rather strongly curved upwards, similar to Emery's
figure of quadridens (Term, fuzet., 1897, 20, pi. 15, fig. 25). The color is dark
fiiscous to piceous instead of ferruginous.

San Crlstoval: Pamua (Type-localit^O- Wai-ai.
Found beneath a stone.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 341

Two workers in the series are abnormal. Each bears on one side
of ihe pronotum a single elevated tooth similar to those on quadridens.
The specimens might be considered as abortive atavistic approaches
to some ancestral form very close to quadridens. On account of the
presence of these spines in a species where they are normally absent
I consider this character of too little importance to be used in sepa-
rating certain of the species into the subgenus Odontomyrmex. Type.
— M. C. Z. 9,174.

68. Pristomyrmex mendanai, sp. nov.

Worl^er. Length 2.25 mm.

Near P. lucidus Emery. Head as broad as long, with slightly rounded
sides, broadly rounded occipital corners and subtruncate posterior border.
CljTJeus flat, carinate at base; anterior border tridentate, the lateral teeth
acute, the median one obtuse. Frontal carinae acute; antenna! scrobe broad
and shallow. Base of mandibles with large obtuse tooth, blades quadriden-
tate. Antennal scapes slightly surpassing occipital corners; funicular joints,
except the first and those forming the club, somewhat broader than long. Eyes
small and slightly convex, situated at ^es of head a little in front of middle.
Promesothorax convex in profile, without pronotal spines and with obtusely
angulate inferior corners; with a broad median impression extending along
the dorsum to declivity of epinotum. Epinotal spines broad at base and
acuminate and curved apically. Epinotal declivity concave and broadly
margined at sides. Petiolar node as long as peduncle, higher than long; high-
est in front; convex above; concave on posterior surface; from above, longer
than broad and narrowed anteriorly, truncate behind. Postpetiole broader
than petiole and slightly broader than long, with straight borders in front
and behind and slightly convex sides. The anteroventral margin acutely
angulate in profile.

Very shining. Mandibles finely punctate. Head and thorax with scattered
and shallow punctures, the spaces between smooth. Petiole, postpetiole, and
gaster sparsely and finely punctate.

Everjrwhere with moderately abundant fine and long pile.

Uniformly brownish red.

Santa Cruz: Graciosa Bay.

Two workers found beneath stones. This species resembles hicidxis
Emery, but is .smaller and differently colored, the promesonotal dorsum
is strongly impressed and the middle tooth on the border of clypeus is
not acute. The closely related P. levigatus Emery has the sides of
prothorax denticulate in front.

342 bulletin: museum of comparative zoology.

69a. RoGERiA STiGMATiCA Emery.

Term, fuzet., 1897, 20, p. 589, g .

Worker. Length 2.25-2.50 mm.
Female (dealated). Length 2.75 mm.

Differing from the worker in the f oUowing characters : — the epinotal spines
are longer. The mesonotum and scuteUum are strongly striated longitudi-
nally. The striae on base of epinotum are stronger. The petiolar node in
profile is narrower and the peduncle a little longer.

Male. Length L75 mm.

Head, excluding the mandibles, a little broader than long, broadest behind,
with rounded corners and straight occipital border. Clypeus convex, anterior
border rounded. Mandibles large, their blades dentate. Antennae 14-jointed,
slender, thickened distally; scape thickened in front of middle, extending
two thirds the distance to occipital corners; funicular joints longer than broad;
terminal joint four times as long as broad and equal in length to the two pre-
ceding joints together. Eyes convex, located at about one fourth their length
from base of mandibles. Ocelli rather flat, situated in large foveae. Mesono-
tum with strong Mayrian furrows; elevated in front. Scutellum broader than
long, convex. Epinotum long, with subequal base and declivity. Petiolar
node low, the anterior surface shorter and more convex than the posterior;
peduncle as long as node, flattened above, sides margined. Postpetiole longer
than high, with a flattened anterior dorsal surface and short convex posterior
surface above, broader than long, broadest in front, with obtusely angulate
anterior corners. First gastric segment somewhat compressed dorsoventrally.
Legs long, not very slender. Wings with long and narrow stigma and a single
large cubital cell.

Somewhat shining. Mandibles and clypeus finely punctate. Front of
head rugosely punctate, posterior portion of head and pronotum finely punc-
tate. Mesothorax and scutellum rugosely striolate; epinotum finely and
densely punctate. Petiole, postpetiole, gaster, and legs finely punctulate and
shining.

Fine, long, and semirecumbent pale hairs throughout, but most abundant on
head, thorax, and gaster.

Color yellowish brown; pronotum, thoracic pleurae, petiole, and postpetiole
lighter. Legs, clypeus, and antennae yellow. Wings hyaline, veins pale,
with dense pale hairs.

Ysabel: Fulakora. Ugi: Pawa.

Occurs in small colonies beneath stones in damp localities. This
species has a curious exudating habit. When the nest is disturbed the

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 343

workers twist the gaster to one side and emit from the anal opening a
cyh'ndrical white mass of viscid matter, often up to ten millimeters in
length, that resembles a small worm. While being emitted, this curls
and twists in a life-like manner so that the first time it was observed
and before I had located the motionless ant, I mistook it for a cluster
of very small worms. This same habit is common also to an un-
determined Fijian species, probably of the same genus. Type. —
M. C. Z. 9,175.

70. ROGERIA EPINOTALIS, Sp. nOV.

Worker. Length 2-2.25 mm.

Head longer than broad, broadest behind, with rounded corners and straight
occipital border. Mandibular blades with five small teeth. Clypeus convex,
rounded anteriorly. Antennal scape extending two thirds the distance to
occipital corners, first funicular joint a little longer than the two succeeding
joints taken together, joints two to six strongly transverse, seven to nine less
strongly so, terminal joint equal in length to the three preceding joints. Eyes
little convex, situated at sides of head at a distance equal to their longitudinal
diameter from bases of mandibles. Promesothorax robust, evenly convex
above and at sides. Humeral angles obtuse. Epinotum broadly and strongly
concave; base very distinctly margined in front and at sides; declivity indis-
tinctly separate from base and more strongly concave; spines a little longer
than their width at base, acuminate, divergent. Petiolar node in profile
deeper than long, not narrowed above, with convex dorsum and slightly convex
anterior and posterior surfaces; from above, twice as broad as long; peduncle
nearly as long as the node. Postpetiole in profile longer than high; anterior
surface short and flat, dorsal and posterior surfaces convex; from above,
one and one third times as broad as long, broadest in front of middle, with
narrowly rounded sides. Legs rather stout.

Shining. Mandibles sparsely punctate. Head rugose and carinate. The
carinae between antennal scrobes subparallel, on vertex more irregular and
becoming somewhat reticulate. Those in scrobes more widely separated and
oblique. Promesonotum with coarse punctures, which are widely separated
in front, very sparse on disc and coarser, becoming confluent posteriorly.
Pleurae coarsely punctate, except the metapleurae which are smooth and very
shining below, with a margin of coarse punctures above. Base and declivity
of epinotum smooth and very shining except for a delicate transverse carinula
between the two surfaces,. Anterior face of petiolar node and peduncle
smooth and very shining; rest of petiole and the postpetiole coarsely punctate.
Gaster, legs, and antennae minutely punctulate and shining.

Head and body, e.xcepting epinotum (which is glabrous), anterior surface of

344 bulletin: museum of comparative zoology,

petiolar node and the peduncle with long silky, flexuous hairs, which are
semirecumbent on posterior portions of gaster and erect elsewhere. Antennae
and legs with shorter hairs.

Color ferruginous; legs, antennae, and mandibles testaceous. Pilosity
' white.

Female. Length 2.25 mm.

OcelU small. Similar to worker. The petiolar node is thicker and the
epinotal spines stronger. Wings hyaline. Veins and stigma brown.

Ysabel: Fulakora (Type-locality). Ugi: Pawa. San Cristoval:
Star Harbor. Malaita: Auki.

Described from a few specimens found in very small colonies beneath
stones. The concave and impunctate epinotum, the stronger spines,
the irregular punctation of the pro- and mesothorax, the smaller size
and different coloration distinguish it from the preceding species.
Tijpe.—M. C. Z. 9,176.

71. Tetramorium salomo, sp. nov.

Worker. Length 2.50 mm.

Head a little broader than long, sides in front of eyes somewhat convergent,
occipital border shallowly concave. Mandibles 6 -dentate, the two anterior
teeth larger than the others. Clypeus subcarinate at middle, straight at ante-
rior border. Frontal ca/inae delicate though distinct, scrobes poorly defined.
Antennal scapes shghtly surpassing occipital corners; funicular joints three to
seven distinctly transverse, the others as long or longer than broad. Thorax
evenly convex in profile, not impressed, humeri narrowly rounded. Epinotal
spines very slender and acute, a little longer than their distance apart at tips,
nearly straight, extending backward and slightly upward. Inferior epinotal
angles produced as flat, triangular spines, rather obtuse at tips and a little more
than one third as long as dorsal spines. Petiolar node from above oval, nar-
rowed in front, from the side longer than broad evenly convex above, in front
evenly rounding into the slender peduncle; anteroventral spine elongate,
rounded at tip. Postpetiole broader than petiole and a little broader than
long. Legs rather slende.'.

Gaster very shining, legs moderately shining, the rest subopaque. Man-
dibles finely punctate, anterior border with short striae. Clypeus with uneven
surface and fine longitudinal striae. Head with a series of serrated, slightly
tortuous striae which have a tendency to become reticulate on occiput, the
interspaces foveolately punctate and rugulose. Thorax and petiole coarsely

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 345

reticulately rugose and punctate between the rugae. Postpetiole more finely
sculptured. Gaster smooth.

T'ilosity long and abundant, erect on head and body, suberect on appendages.

Color brownish yellow with the tip of gaster infuscated.

Malaita: Auki.

Described from one worker.

This species resembles mdleum Emery from New Guinea in general
structure and in the long and slender epinotal spines, but the latter
species has the petiole in profilie much more narrowed above, the post-
petiole distinctly broader than long and broader than the petiole;
in salomo the postpetiole is not broader than long and the sculpture of
the body is very different.

72. Tetramorium melanogyna, sp. nov.

Worker. Length 2.5 mm.

Head longer than broad, about as broad in front as behind, with very feebly
convex sides and shallowly ^joncave occipital border. Mandibles with three
strong teeth on apical half and three weak ones lateral^. Clypeus evenly
convex, anterior border straight. Frontal carinae rather strong, little diver-
gent, extending nearly to occipital margins and bordering distinct, though
shallow scrobes. Antennal scapes extending about four fifths the distance to
occipital corners; first funicular joint as long as the three following joints

Fig. 28. — Tetramorium melanogyna Mann. Worker. Lateral view of thorax and petiole.

together, joints two to eight transverse; terminal joint a little longer than the
two preceding joints together. Eyes situated at middle of sides of head.
Thorax, seen from above, with subangulate humeri, constricted between meso-
and epinotum; in profile, with shallow mesoepinotal impression. Epinotal
spines nearly straight, longer than their distan.^e apart at base, directed upward
and backward; episternal spines thick basally, with the apical half narrow and
the tips acute. Petiole with a distinct peduncle, a little longer than the node;
the node in profile as high as long and highest at posterior margin, with sloping

346 bulletin: museum of comparative zoology.

anterior face and moderately convex dorsal surface, seen from above it is
elongate oval and somewhat narrowed in front. Postpetiole subglobular,
slightly broader than long and distinctly broader than the petiole. Femora
rather strongly thickened.

Shining throughout. Mandibles finely striate. Disc of clypeus with three
longitudinal rugae, which continue up the front parallel to an outer pair of
frontal rugae and become retulate on vertex and occiput, with the spaces
between finely and shallowly punctate. Cheeks and sides of occiput reticu-*
lately rugulose, antennal foveae bordered with a ridge that is distinctly angu-
late at middle of sides of the pro- and the mesonotum. Petiole and post-
petiole more finely reticulate, with the sides of the former and especially of
the peduncle strongly margined. Gaster with short, parallel striae at base.

Color light brownish yellow with the gaster infuscated.

Feviale. Length 2.80 mm.

Epinotal spines shorter and stouter than in the worker; episternal spines
more slender.

Sculpture of head as in worker. Mesonotum and scutellum Aith longi-
tudinal, somewhat irregular, punctate rugae between. The /est similar to
worker.

Color dark brown to black. Wings clear, with pale veins and stigma.

Ugi: Pawa (Type-locality). Three Sisters: Malapaina.

The one colony that I found was in a hollow twig, and contained
two of the curiously dark-colored females.

This species closely resembles pacifirum Mayr, but has the petiole
less compressed in front. The coloration, yellow-brown, with infus-
cated gaster, is different.

73. Tetramorium guineense (Fabricius).
Formica guineense Fabr., Entom. syst., 1793, 2, p. 357, y .

Malaita: Auki.

The finding of only one colony of this species indicates that it is not
especially abundant in the islands, though no doubt it will eventually
become widespread in the plantation districts.

74a. Tetramorium scrobiferum Emery, var. elysii, var. nov.

Worker. Length 2.75-3 mm.

Head longer than broad, broadest behind eyes; sides in front of eyes nearly
straight, occipital corners rounded, border shallowly concave. Mandibles

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 347

with five short, subequal teeth. Clypeus moderately convex, bluntly carina te
at middle; anterior border almost straight at middle. Antennal scrobes well
defined, extending beyond eyes, the carinae bordering the inner sides somewhat
crenulate and narrow but conspicuously stronger than the carinae of front and

Fig. 29. — Tetramorium scrobiferum elysii Mann. Worker. Lateral view of head, thorax,
and petiole.

vertex. Antennae short and thick; scapes extending less than three fourths
the distance to occipital corners; fimicular joints two to eight twice as broad
as long; terminal joint about as long as the two preceding joints. Eyes not
very convex, located in front of sides of head a little behind the middle. Thorax
robust, the sides nearly straight; all sutures obsolete; prothorax margined at
sides, with angulate humeri. Epinotum with base and declivity rounding
into each other, the latter flat: spines acute, curving downward, moderately
divergent, a little shorter than their distance at base. Petiolar node higher
than long, with short pedicel which is angulate at anterior corners. From
above globose, as long as broad. Postpetiole broader than petiole and broader
than long.

Subopaque. Mandibles punctate and striate, the striae stronger basally.
Clypeus with two carinae lateral to the median one. Antennal scrobes opaque
and very densely punctate; head finely and reticulately carinate, the inter-
spaces punctate and feebly shining. Thorax and epinotum densely punctate.
Petiole and postpetiole rugulose and more shining. First gastric segment
uniformly, shallowly, and very densely punctate and opaque; terminal seg-
ments punctulate and rather more shining. Antennae punctate. Legs shin-
ing.

Pilosity erect, sparse and white in color.

Color ferruginous. Legs lighter. ,

Three Sisters: Malapaina (Type-locality). Santa Anna.

The workers on which the description is based agree closely with
Emery's description of the typical form from Berlinhafen, New
Guinea, except in the lighter color. The epinotal spines in my speci-
mens are proportionately longer than shown in the drawing of scrobi-
ferum (Term, fuzet., 1897, 20, pi. 15, fig. 31). Typc.— M. C. Z. 9,177.

As Emery has pointed out, this curious species approaches Calypto-
myrmex in the structure of the antennal scrobes, the shortly peduncu-
late petiole, and the short and thick antennae.

348 bulletin: museum of comparative zoology.

75. Tetramorium tonganum Mayr.

Verh. K. K. zool.-bot. gesellsch. Wien., 1870, 20, p. 972, 976, y .
SanCristoval: Wainoni Bay. Guadalacanar: Aula.

76a. XiPHOMYRMEx costatus Emery, subsp. flavescens Emery.

Term, fuzet., 1897, 20, p. 588, ^ .
Male. Length 3 mm.

Head, excluding eyes, longer than broad; occipital border straight, the
angles broadly rounded. Mandibles similar to those of worker, with strong
apical and subapical teeth. Clypeus convex, carinate at middle, broadly
rounded in front. Eyes very large and convex, located at a distance equal to
one fifth their length from bases of mandibles; ocelli large. Antennae slender,
10-jointed, their scapes surpassing occipital corners; first funicular joint one
and one half times as long as broad, second joint longer than the third and
fourth together, joints three to eight subequal, cyhndrical; terminal joint one
and one half times as long as penultimate. Mesonotum rather flat at apical
half, without Mayiian furrows. Scutellum transverse, httle convex above.
Epinotum deeply transversely grooved at base; the rest flat, narrowly mar-
gined at sides. Petiole long and slender; petiole as long as node, flattened
above and narrowly margined at sides; node longer than high; from above
longer than broad, transversely impressed apical ly Postpetiole rounded,
broader than long. Legs long and slender.

Shining, petiole, postpetiole, and gaster more so than the rest. Mandibles
finely strigose. Clypeus with two lateral carinae which are much stronger
than the median one. Head, thorax, and epinotum with reticulate carinae.
Petiole, postpetiole, and gaster smooth. Legs finely and rather densely,
though shallowly punctate. Antennal scapes finely, funiculus rugosely punc-
tate.

With sparse, scattered, semicircumbent hairs throughout. Color testaceous;
funiculus, tibiae, and tarsi fuscous. Wings hyaline; veins and stigma fuscous.

Bio. Three Sisters : Malapaina. San Cristoval: Pamua. Malaita:
Auki, Interior of. Florida: Tulagi. Russell: West Bay. Ysabel:
Fulakora.

The form that I refer to this subspecies is common in the forests,
where workers are frequently seen on tree-trunks and leaves. The
colonies are small and live in hollow twigs (in most instances rotten
ones), beneath bark and more rarely in the ground.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 349

77. XiPHOMYRMEX BISMARCKI Forel.

Mitt. Mus. zool. Berlin, 1901, 2, heft 1, p. 17, ^ .

New Georgia: Rubiana Lagoon.

Two workers were found on the ground near the beach.

This species is distinguished from the preceding by its smaller size
and the shape of the head, which is shorter, with more convex sides
and with the occipital border more concave. The eyes are smaller.
The color is reddish brown, except the petiole, postpetiole, and gaster
which are brownish yellow.

78. Triglyphothrix obesa (Em. Andre).

Tetramorium obesum'Em. Andre, Rev. entom., 1887, 6, p. 294, ^ .
Male. Length 2.75 mm.

Head, excluding eyes, a little longer than broad, broadest behind eyes,
posterior angles obtuse, border straight. Mandibles large and well formed,
with large, triangular apical and subapical teeth. Clypeus convex, anterior
border feebly rounded. Antennae robust, 10-jointed; scapes extending
beyond posterior borders of eyes; first funicular joint globose; second joint
two and one half times as long as broad and longer than the two succeeding
joints together, third joint small; joints four to eight longer than broad, sub-
moniliform; terminal joint slightly shorter than the two preceding joints
together. Eyes large and convex, located at a distance of one fourth their
longitudinal diameter from anterior corners of head. Ocelli large. Thorax
robust, with strongly impressed Mayi'ian furrows. Scutellum little convex
basally, margined at sides; spines reduced to broadly obtusely angulate lamel-
lae. Petiolar node longer than high; from above subglobose and shghtly
broader than long. Postpetiole from above similar to petiole but broader.
Legs long and slender.

Gaster and legs shining, the rest more feebly so. Mandibles punctate.
Head irregularly striate, the striae longitudinal on front and clypeus, transverse
between eyes and frontal carinae and oblique and somewhat reticulate on
vertex and occiput.

Pronotum coarsely punctate, especially at sides. Mesonotum reticulately
striate. Scutellxma punctate. Epinotum and petiolar peduncle carinate.
Petiolar node and postpetiole coarsely but shallowly punctate. Gaster and
legs finely punctate.

Pilose everywhere, similarly to worker.

350 bulletin: museum of comparative zoology.

Color testaceous throughout. Wings hyaUne, veins and the broad stigma
yellowish.

Ugi: Pawa.

The many workers before me, agree closely with Andre's descrip-
tion and with workers from India received from Dr. Forel. This
species, like the related species siriatidcns Emery which Forel has
recorded from the Bismarck Archipelago but which I failed to find in
the Solomons, appears to be extending its range.

79a. Triglyphothrix fulviceps Emery, subsp. antennata, subsp.

nov.

Worker. Length 2.25-2.50 mm.

Head a little longer than broad, about as broad in front as behind, with
convex sides, narrowly angled posterior corners and concave border. Mandi-
bular blades 6-dentate, the three basal blades short, the others longer and acute.
Clypeus rather strongly convex, sharply carinate at middle; anterior border
feebly rounded. Antennal scapes almost attaining occipital corners; funicular
joints two to eight slightly transverse; terminal joint longer than the two
preceding joints taken together. Eyes large and convex, located at sides

Fig. 30. — Triglyphothrix fulviceps antennata Mann. Worker. Lateral view of thorax
and petiole.

of head, a little in front of middle. Thorax robust, in profile evenly convex ;
from above, broad in front with rounded humeral angles and nearly straight
sides. Base and declivity of epinotum not distinct; spines divergent, acute,
about as long as distance between their bases. Petiolar node in profile as long
as high, nearly straight in front, convex above and behind; from above, a little
longer than broad, narrowest in front, with convex sides. Postpetiole trans-
versely ovate; broader than petiole.

Somewhat shining, the gaster and legs more so than the rest. Mandibles
and clypeus coarsely striate. Head longitudinally striate, the striae becoming
reticulate on vertex, sides, and occiput. Thorax, epinotum, petiole, and post-
petiole coarsely, cribrately punctate. Gaster coarsely but shallowly punc-

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 351

tate basally, more sparsely behind. Everywhere with abundant short silky
hairs.

Head and legs, except part of femora, and antennae brownish yellow.
The rest dark reddish brown, with the basal half of first gastric segment lighter
than the other parts.

Santa Cruz: Graciosa Bay.

Described from specimens taken from a small colony beneath a stone
and from several isolated workers.

Very close to typical fuhiccps but the antennal scapes extend dis-
tinctly more than three fourths the distance to occipital corners, the
petiolar node is not broader than long and the color is lighter.

80. Triglyphothrix mayri, sp. nov.

Worker. Length 1.75-2 mm.

Head a little longer than broad, as broad in front as behind, with slightly
convex sides, rounded corners and moderately concave occipital borders.
Mandibles with five short teeth, the apical and subapical longer than the others.
Clypeus convex, carinate at middle, slightly rounded at anterior border;
Antennal scapes extending a little more than three fourths the distance to
occipital corners; funiculus stout, joints two to eight strongly transverse,
terminal joint longer than the two preceding joints taken together. Eyes
large and convex, located at sides of head at a distance equal to their longi-
tudinal diameter from anterior corners. Thorax robust, evenly convex above.
Epinotal spines triangular, rather stout, about as long as broad at base: epi-
sternal spines broader than long. Petiolar node in profile longer than high;
the pedicel much shorter than node, from above, subglobose about as long as
broad. Postpetiole a little broader than petiole and a little broader than long.
Legs stout.

Subopaque, except mandibles, apical segments of gaster and legs which are
shining. Mandibles punctate and indistinctly striate. Clypeus longitudi-
nally striate. Head, thorax, petiole, and postpetiole densely and coarsely
punctate, the thorax and epinotum faintly and irregularly carinulate. Epi-
notal declivity smooth. First gastric segment with large but shallow punc-
tures over entire dorsum.

Densely covered with fine, short, white pilosity.

Color brownish yellow; gaster darker than the other parts.

San Cristoval: Star Harbor.

This species approaches walshi Forel from British India but differs
in the shorter petiole to the first node, in the head being as broad in

352 bulletin: museum of comparative zoology,

front as behind, in the finer striation of the mandibles, and in color.
From the closely related viusculus Forel mayri differs in the shape of
the petiole which is not transverse. The punctation of the entire
dorsum of the first gastric segment is distinctive.
Typc—M. C. Z. 9,178.

81. Triglyphothrix pulchella, sp. nov.

Worker. Length 1.75-2 mm.

Head longer than broad, little broader behind than in front, with feebly
convex sides, rounded posterior corners and shallowly concave border. Mandi-
bles 6-dentate, the three basal teeth indistinct. Clypeus convex, carinate at
middle, broadly rounded in front. Antennae stout, their scapes extending
three fourths the distance to occipital corners, funicular joints two to eight
transverse, terminal joint longer than the two preceding together. Thorax

Fig. 31. — Triglyphothrix pulchella Mann. Worker. Lateral view of thorax and petiole.

robust, humeral angles narrowly rounded. Epinotal spines slender, longer
than the distance between them at base and strongly divergent. Petiolar
node a little broader than long; pedicel short. Postpetiole a little broader
than long and only slightly broader than the petiole.

Subopaque, except terminal half of gaster which is shining. Mandibles
somewhat shining, punctate, and striolate. Head, thorax, epinotum (except
declivous portion), petiole, and postpetiole coarsely and densely punctate.
Basal half of first gastric segment punctate and longitudinallj' striate, the rest
finely punctate. Legs and antennae somewhat shining and finely punctate.

Pilosity short, white, and abundant.

Color pale brown; legs, antennae and mandibles lighter, head yellow.

Female (dealated). Length 2.25 mm.

Similar to worker. The ocelli are small. The mesothorax and scutellum
are rather flat and longitudinally carinate at middle. The striation at base
of gaster is stronger than in the worker.

Malaita: Auki.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 353

The very long, slender, and divergent spines readily distinguish this
species. Type. — M. C. Z. 9,179.

Key to Papuasian Species.

Petiolar node distinctly longer than broad (Indo Malaya; Australia;

West Indies; North America, Louisiana) striatidens Emery.

Petiolar node as broad or broader than long 1 .

1. Larger species (length 2.5-3 mm.); mandibles not striate; gaster smooth

and shining. (India; Solomons) obesa Em. Andre.

Smaller species (Length 1.75-2.50 mm.), mandibles striate; gaster punc-
tate, at least basally 2.

2. First gastric segment not striate basally, coarsely but evenly punctate
throughout; epinotal spines very short. (Solomons). .. .ma?/ri Mann.
First gastric segment striate and punctate basally; epinotal spines

longer 3.

3. Epinotal spines long, slender and strongly divergent; color pale brown,

with head yellow (Solomons) pidchella Mann.

Epinotal spines not longer than distance between them at base; color
dark reddish brown to black, with head lighter 4.

4. Petiolar node broader than long; antennal scapes extending three fourths

the distance to occipital corners ; color black, with head and legs reddish

brown. (New Guinea) fulviceps Emery.

Petiolar node not broader than long; antennal scapes almost attaining
occipital corners; color dark reddish brown, head and gaster lighter.
(Santa Cruz) fulviceps subsp. antennatus Mann.

82. Strumigenys godeffroyi Mayr.
Sitzungsb. Akad. wiss. Wisn, 1888, 53, p. 516,^ .

Santa Cruz: Graciosa Bay. San Cristoval : Pamua, Wai-ai, Wain-
oni Bay. Three Sisters : Malapaina. Ugi: Pawa. Malaita: Auki.
Ysabel: Fulakora.

This is a very common species. It nests beneath bark or stones or in
rotten wood and has been found in Termitaria.

83. Strumigenys decollata, sp. nov.

Worker. Length 4 mm.

Form long and slender. Head twice as long as broad, twice as broad at
occiput as at clypeus, with narrowly rounded occipital corners and strongly

354

bulletin: museum of comparative zoology.

impressed occipital border; sides convex in occipital region, then evenl}^ con-
cave to and a little past the eyes, and then convex to the clypeal
border. Clypeus flattened at base, acuminate behind; in front excavated
broadly and triangularly, so that the sides form rectangular lobes. Antennal

foveae well defined and margined as far as the eyes,
then poorly defined and without margin. Scapes
stout, extending less than two thirds the distance to
occipital corners, about as long as mandibles; first
funicular joint a little longer than the second and
third, which are subequal and two and a half times
as long as broad; terminal joint rather stout, one
and a half times as long as the penultimate.

Mandibles long and arcuate, of equal thickness
from base to near apex; the inner border a little
before the middle has a small, but distinct triangular
tooth and anterior to this a second, smaller tooth;
midway between this and the terminal tooth is a
slender tooth, more than half as long as the terminal
ones; upper terminal teeth long and slender, the
lower ones much shorter than the others and strongly
twisted. Thorax long and slender, the sutures obso-
lete. Promesothorax narrowed anteriorly into a neck,
which is feebly margined; broadest behind middle,
with convex surface and rounded sides, evenly sloping
behind to the epinotiun. Epinotum in profile flattened
basally, rounded at sides; armed with two stout trian-
gular spines, which are about as long as the very short
declivity. Peduncle in profile shorter than the node,
which is nearly twice as long as high and feebly convex above; seen from above
more than twice as long as broad, the sides subparallel to near the apex, then
divergent. Postpetiole from above a little broader than long, campanulate in
shape, broadest behind, narrowed in front, rounded above and at sides. Gastcr
sliort and stout. Legs Ions and slender.

Fig. 32. — Strumigenys de-
collata Mann. Worker
Front view of head.

Fig. 33. — Strumigenys decollata Mann. Worker.

Mandibles shining, finely punctate. Clypeus shining, with scattered
punctures. Head feebly shining, densely and reticulately punctate; vertex
more finely than the rest. Thorax more shining, with widely separated,
strong punctures. Base of epinotum smooth and very shining. Petiole

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 355

shining, with moderately coarse punctation. Postpetiole more shining, with
sparser punctation. Gaster very shining, finely and sparsely punctate. Legs
somewhat shining and strongly punctate.

Occipital margin, mandibles, and tip of gaster with very sparse and fine
erect pile. Scapes with a row of ten to twelve stout curved hairs, funiculus
with appressed and glistening hairs. Legs with sparse, appressed hairs.
Petiole and postpetiole with fungiform processes in a narrow band at the
apical margins and more abundantly on the ventral surface. The greater
part of the body is without erect pile or pubescence.

Color reddish brown; pilosity white.

Male. Length 2.25 mm.

"Head, excluding eyes, longer than broad; sides rounded; occipital border
shallowly concave between the posterior ocelli. Cheeks a third as long as
the eyes. Clypeus convex basally, anterior border feebly concave. Mandibles
short; bidentate at apex. Eyes convex, their longitudinal diameter as long
as their distance from the posterior ocelli. Antennae 13-jointed; scape less
than twice as long as it is broad at base; first funicular joint as thick as the
scape and more than half as long; remaining joints very gradually becoming
longer toward apex; terminal joint' one and a half times as long as the penulti-
mate. Pronotum flattened and collar-like in front. Mesonotum rounded
above; with strong Mayrian furrows. Scutellum broader than long, rather flat
in front, convex behind. Epinotum flattened basally; its very distinct declivous
portion flat and broadly margined at sides. Petiole in profile rounded above,
much longer than high; from above, twice as long as broad. Postpetiole a
little longer than broad, rounded above and at sides; in profile a little longer
than deep. Legs slender.

Head and thorax subopaque; very densely punctate. Petiole and post-
petiole punctate at sides; above smooth and shining. Gaster very shining.
Legs subopaque, with fine but dense punctation.

Antennae with abundant, glistening white oppressed hairs, as in the worker.
Legs with a few fine hairs which are more abundant on the tarsi. The rest
of the body is without pilosity except for a very few hairs on the thorax and
apically on the gaster.

Color black ; the head and thorax intensely so, the rest dilute.

Wings hyaline. Veins and stigma brown.

Ysabel: Fulakora.

84. Strumigenys antaeus, sp. nov.
Worker. Length 3.25 mm.

^ft'

Near S. chyzeri Emery.

Head broadly cordate, deeply excavated behind, occipital corners narrowly
rounded, sides strongly constricted in front of eyes; cheeks swollen basally,

356

bulletin: museum of comparative zoology.

but not angulate. Cljrpeus flattened at base; convex in front, end divided
into two lobes by the strong, triangular medial incision. Mandibles in length
about equal to the distance from their base to the eyes ; thickened in front of
middle as in chyzeri and armed with two long teeth at apex and a short and
curved tooth situated dorsally on the swollen position and not extending to
the inner mandibular margin. Antennal scapes extending more than half the

distance to occipital corners; funicular joints
two and three scarcely longer than broad;
fourth joint as long as the three preceding
joints. Eyes little convex, composed of about
twelve ommatidia. Prothorax flat above;
margined at apical third of sides; humeral
angles rounded, the disc with irregular carinae.
Promesonotal suture marked by a carina.
Mesonotum flattened basally, then declivous
to the feeble mesoepinotal impression; mar-
gined at sides. Epinotum with subequal base
and deJclivity; base flattened, twice as long as
broad and indistinctly margined at sides, de-
clivity convex, twice as long as broad and
sharply margined at sides; spines stout, half
as long as base of epinotum. Peduncle of
petiole longer than the node; node in profile
rounded above; from above, longer than broad, with a margin of spongiform
appendages behind. Postpetiole transverse, with a narrow border of spongi-
form appendages in front, on sides and behind. Legs long and rather stout.

Mandibles shining and rather closely punctate. Head, thorax, epinotum,
and petiole feebly shining, granuloscly punctate and with reticulate carinulae
which are most abundant on the head, and lacking on the epinotum and
petiole. Postpetiole and gaster more shining and densely punctulate. Legs
somewhat shining, densely punctate. Head and body with scattered, erect,
and strong pile; first gastric segment costate at base; pile of scapes and legs
semierect and very strongly curved.

Color brownish yellow; mandibles darker.

Fig. 34. — Strumigenys antaeus
Mann. Worker. Front view
of head.

Ysabel: Fulakora.

A number of workers were found beneath leaves on the ground and
a colony in rotten wood. The deeply excavated clypeus readily
distinguishes antaeus from chyzeri Emery which has the anterior
border arcuate. The two species are closely related. Strumigenys
loriac Emery from New Guinea has the short mandibular spine before
the two terminal ones, situated dorsally, and the cheeks are markedly
angulate midway between the eyes and the clypeus. Type. — M. C. Z.
9,180.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 357

84a. Strumigenys antaeus Mann, var. fuscior, var. nov.

Worker. Length 2-2.25 mm.

Differing from the preceding in its smaller sme and in coloration, being
reddish brown, with the gaster, except the apex much darker than the rest of
the body.

San Cristoval: Pamua. Ugi: Pawa. Malaita: Auki.

Szabo (Ann. Mus. nat. Hung., 1910, 8, p. 368) states that Biro found
in New Guinea large and small workers of S. loriae Emery, in the same
nest, indicating polymorphism in the species. The specimens that I
have separated from aniacus as the variety fuscior certainly are very
similar in structure to the type form, but they were found always in
separate colonies, the larger and smaller forms never together. I have
seen no specimens intermediate in size between the two.

85. Rhopalothrix isabellae, sp. nov.

Worker. Length 3 mm.

Head about as long as broad; strongly excavated behind and with obtusely
angulate corners; oblique portions of sides with straight margin. Clypeus
narrowed behind and broadly and shallowly concave in front; the suture
separating it from the front shallow and rather poorly defined. Mandibles
strongly constricted at basal third and much more slender than in the related
species. Antennal scapes at base about a third as
broad as long; funicular joints two _ and three only
slightly transverse, joints three and four longer than
broad, terminal joint longer than the four preceding
joints together. Pronotum shallowly impressed at
middle and obtusely gibbous on either side; humeral
angles rounded, without any indication of tubercles.
Promesonotal suture obsolete. Mesonotum slightly
convex in front, and depressed behind. Mesoepinotal
suture distinct. Base of epinotum strongly impressed ^ „^ r.. , ,. ■ •

, ^ . o J I pjg gg — Rhopalothrix isa-

mesally, marginate at sides of impression, separated bellae Mann. Worker,
from the concave declivous portion by an acute margin; Front view of head,
spines short and obtuse apically, their bases extended

as narrow lamellae along borders of declivity. Petiolar node longer than broad,
slightly broadest apically and very feebly convex at sides and front; flat above;
about as long as peduncle. Postpetiole one and a third times broader than
long, narrowed in front, sides little convex; dorsal surface with a faint medial
impression which terminates in front in a broad and shallow fovea; strongly

358 bulletin: museum of comparative zoology.

impressed at apex and elevated into an obtuse tubercle on either side of
impression. First gastric segment indistinctly impressed at middle of base.

Mandibles shining and punctate. Head feebly shining. Cl^-peus strongly
and foveolately punctate; front reticulately and rugosely punctate; vertex
with two large shallow foveae and cribriform punctures between. Pro-
mesothorax coarsely and reticulately costate and subopaque. Epinotum at
sides more finely costate; impressed portion foveolately punctate; declivity
smooth and shining. Petiole finely rugose. Postpetiole and gaster more
shining and uniformly cribrate. Antennae and legs coarsely and densely
subopaque.

Fine, squamiform hairs on head and antennal scapes, coarse ones on legs.
Clavate hairs on legs, outer border of scape and one in each of the two foveae
on vertex; fine hairs on funiculus, tarsi, and ventral surface of gaster.

Fuscopiceous; legs, mandibles, antennae, and lamellate borders of epinotum
fuscorufous. Hair white.

Ysabel: Fulakora.

The single worker of this anomalous species was found beneath a
log in the forest. The curious form of the mandibles and the relatively
long postpetiole readily distinguish isabcllae from the other Papuasian
species. The lack of clavate hairs on the dorsal surface is probably
not due to their being rubbed off, for such hairs are abundant on the
legs.

86a. Rhopalothrix procera Emery, subsp. malua, subsp. nov.

Male. Length 3 mm.

Head, excluding eyes, much longer than broad; sides posterior to eyes
parallel, then oblique to the occipital border which is shall owly concave.

Vertex broadly and deeply impressed between the
ocelli. Frontal cannae strongly elevated. Clypeus
convex; concave anteriorly and angulate at sides.
Mandibles small and acuminate apically. Antennae
slender, scape broad basally, twice as long as first
funicular joint; first funicular joint about a third
as long as the second; joints two to eleven elongate
and cylindrical, becoming shorter toward apex;
terminal joint one and a half times as long as
penultimate. Eyes shorter than border of cljrpeus,
very convex. Mesothorax flattened posteriorly
Fig. 36. - Rhopalothrix ^^^ impressed at middle. Scutellum about as

procera malua Mann. i i ^•-'^^^ u.a

Worker. Front J view ^^^S ^^ broad, and rather flat above. Epinotal
of head. spines rounded apically. Petiolar node as long

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

359

as broad; impressed at middle of posterior border; antero ventral spine
long and slender. Postpetiole much broader than long; with a short longi-
tudinal impression at middle of apex.

Fig. 37. — Rhopalothrix procera malua Mann. Worker. Lateral view of thorax,
petiole, and abdomen.

Opaque, except gaster and legs which are very faintly shining. Densely
and coarsely punctate throughout, with the punctation of gaster and legs
somewhat more shallow than the rest. Above wdth sparse, semiclavate hairs
and scattered fine and recumbent pilosity, ventral surface and coxae with fine,
long, and erect hairs. Legs and antennae with recumbent pilosity.

Fig. 38. — Rhopalothrix procera malua Mann. Male. Lateral view.

Black; mandibles, antennae, lamellate extensions of epinotal spines, apex
of gaster and legs, except coxae, fuscoferruginous.

Wings (length 3.5 mm.) strongly infuscated; veins brown.

Santa Cruz: Graciosa Bay. San Cristoval: Pamiia, Wainoni Bay,
Wai-ai. Three Sisters: Malapaina. Ysabel: Fulakora. Rendova.

360 bulletin: museum of comparative zoology.

A large series of workers from several localities agree with Emery's
description of yroccra except that the postpetiole is not margined at
sides and the size averages smaller. (Length 3-3.5 mm.)

The female is distinctly smaller (Length 4 mm.) than in typical
proccra, though agreeing in other respects. The wings are rather
strongly infuscated, with brown veins.

Small colonies were found between layers of rotten wood or beneath
bark when there was moisture. It is a very slow moving species.

There is considerable variation in the color of the workers, from
rufotestaceous to rufofuscous, though none approach the very dark
<;»li>t of the following variety. Typc.— M. C. Z. 9,18L

86b. Rhopalothrix procera Emery, subsp. malua Mann, var.

melanotica, var. nov.

A single worker from Tulagi is different from any of the large series
of the preceding form before me in being piceofuscous in color, much
darker than the most extreme of subsp. malua.

Key to Papuasian Species.

Smaller species, less than 2 mm. in length L

Larger, more than 2 mm. in length 3.

1. Petiolar node not longer than broad; front with a transverse impression

between clypeus and vertex (New Guinea) hrevicornis Emery.

Petiolar node twice as long as broad; front without transverse impres-
sion 2.

2. Head broader; very shallowly and broadly excavated behind. (New

Guinea) -punctata Szabo.

Head narrower, more deeply and narrowly excavated behind. Head,
thorax, and abdomen with clavate and squamiform hairs intermingled.
(New Guinea) mixta Szabo.

3. Oblique portions of sides of head convex in outline; occipital border feebly

concave; supraocular portion of head with squamiform scales. Length

2.3 mm. (New Guinea) hiroi Szabo.

Oblique portions of sides of head slightly concave in outline; occipital
border strongly concave; hairs on front uniformly distributed 4.

4. Outer border of mandibles strongly concave before middle, so that the

apical portions are slender. Humeral angles evenly rounded. (Solo-
mons) isabellae Mann.

Mandibles of usual form, not constricted. Humeral angles subgibbous 5.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

361

5. Postpetiole margined at sides. liength 3.5-4.5 mm procera Emery.

Postpetiole not margined at sides. Length 3.-3.5 mm 6.

6. Color fuscoferruginous (Solomons) procera subsp. tnaJua Mann.

Color piceous. (Solomons), .procera subsp. rnalua var. melanotica Mann.

DOLICHODERINAE.

87. TURNERIA PACIFICA, Sp. IIOV.

Worker. Length 2.25 mm.

Head a little longer than broad, nearly as broad in front as behind, with
convex sides and very shallowly concave posterior border. Mandibles with
six teeth, the third and fifth smaller than the others. Anterior border of
clypeus almost straight except at sides, where it
is concave; posterior border broadly rounded.
Frontal area obscure. Frontal carinae very feeble,
extending a little beyond anterior border of eyes.
Eyes elongate, large, nearly a third as long as
head; situated on sides of front, a little posterior
to middle. Antennal scapes slender, almost at-
taining occipital corners; funicular joints longer
than broad. Pronotvun slightly broader than
long, surface rather flat; hmneri with an elevated,
elongate process, like a short carina. Mesonotum
longer than broad. Mesoepinotal impression broad and deep. Base of epi-
notum convex, subtriangular, truncate behind; in profile the posterior mar-
gin appearing bluntly cone-shaped and extending over the short and very

Fig. 39. — Turneria pacifica
Mann. Worker. Front
view of head.

Fig. 40. — Turneria pacifica Mann. Worker. Lateral view of thorax and petiole.

concave declivity. Scale twice as high as long, convex in front, straight
behind, rounded above. Caster short, oval.

Occiput and gaster shining, the rest subopaque; very finely punctate
throughout; and microscopically pruinose. Clypeus and front with a few
erect hairs.

Gaster black, the remainder yellow-brown; terminal joint of funiculus
darker.

362 bulletin: museum of comparative zoology.

Santa Cruz: Graciosa Bay.
Described from a single worker.

This is the fourth species of Turneria. Nothing is known regarding
the habits of these interesting ants.

Key.

1. Eyes smaller, hemispherical. Thora.x margined. Smooth and shining

color black. Length 2-8 (AustraHa) frenchi Forel.

Eyes large and fiat. Thorax not margined, or only pronotum with short
margin. Subopaque species 2.

2. Funicular joints as broad as long; antennal scapes extending only slightly

beyond border of eyes. Mesonotum twice as long as broad. Reddish
brown, gaster blacldsh, funiculus brown. Length 2.5-2.7 mm. (Queens-
land) ^ bidentata Forel.

Funicular joints longer than broad; antennal scapes almost attaining
occipital borders of corners of head. Mesonotum less than twice as
long as broad 3.

3. Pronotum much broader than long. Black; legs and funiculus dark brown,
. mandibles, front of head, scapes, and tarsi yellow. Length 2.2-2.3 mm.

(Bismarck Archipelago) dahli Forel.

Pronotum but little broader than long. Yellow-brown; gaster black.
Length 2.25 mm. (Santa Cruz) pacifica Mann.

88. Iridomyrmex myrmecodiae Emery.

Iridomyrmex cordatus var. myrmecodiae Emery, Ann. Mus. civ .stor. nat. Gen-
ova, 1887, 24, p. 249, ^ .

Santa Cruz: Graciosa Bay. Santa Anna. Three Sisters: Malapaina.
Malaita: Auki. Ysabel: Fulakora. Florida: Tulagi. San Cristoval:
Wai-ai.

This is one of the most abundant ants in the Solomons. It nests
sometimes beneath bark or in crevices on standing trees, but usually
in bulbs of an epiphyte, Myrmecodia sp. {HI. guppyanum), which
grows on the branches of several species of trees and is especially
common on a lowland-inhabiting species of Barringtonia. It has been
shown that Myrmecodia can thrive without the presence of ants, but
I am sure that few of this species do, for among the many that I cut
open, none were without them. Even very young bulbs, less than an
inch in diameter, contained incipient colonies.

There is considerable variation in color among the very large series
before me. The Santa Cruz examples and some from Wai-ai have the
gaster jet black; in the majority of the specimens it is more dilute.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 363

89a. Iridomyrmex anceps Roger, subsp. papuanus Emery.
Ann. Mus. civ. stor. nat. Geneva, 1897, 38, p. 572, ^ .

New Georgia. Rubiana Lagoon.
Three workers agree closely with Emery's description.
Two additional species of Iridomyrmex among my material are
represented only by worker minors.

90. Iridomyrmex rufoniger (Lowne) var.?
Formica rufonigra Lowne, Ent., 1865, 2, p. 279.

Malaita: Auki.

Several workers were found on trunks of trees near the beach.

91. Iridomyrmex dimorphus Viehmeyer.

Abh. berichte K. zool. anthr.-ethn. mus. Dresden, 1912, 14, no. 1, p. 7, pi. 1,
fig. 7, 7a, 7b, ^ .

Worker. Length 4-6 mm.

Head a little longer than broad, moderately narrowed in front, posterior
corners narrowly rounded, border deeply concave. Mandibles stout, with a
larger blunt apical tooth and four to eight smaller teeth. Clypeus slightly
convex, anterior border straight. Frontal carinae very short. Antenna!
scapes barely attaining occipital corners; first funicular joint a little longer
than the second; second joint one and two thirds as long as third, remaining
joints gradually increasing in thickness. Eyes large, not very convex, situ-
ated in front of sides anterior to middle. Pronotum broader than long, evenly
rounded above and at sides. Mesonotum elongate oval, twice as long as broad.
Mesoepinotal suture broadly impressed, with prominently tuberculate spir-
acles. Base of epinotum rounded, twice as long as broad; a little longer than
the declivity into which it broadly rounds. Petiolar node thick, in profile
triangular above, sides narrow. Legs very long and slender.

Shining. Mandibles rugulose. Finely punctate and covered with prui-
nose pubescence, which is most dense on gaster and thoracic pleurae. Rather
stiff erect hairs scattered on head, body, and appendages.

Black. Mandibles and funiculus brown.

Malaita: Interior near Gwountafu. Ysabel: Fulakora.

364 bulletin: museum of comparative zoology.

Described from numerous workers that were travelling across a
path.

Workers were kindly compared by Professor Emery with cotypes,
who writes that they are identical with the smaller workers of dimor-
phus, though the largest of the type lot is much longer (9 mm.) than
anv that I found.

92. Technomyrmex albipes (F. Smith).
Formica (tapinoma) albipes Smith, Journ. proc. Linn. soc. Zool., 1862, 6, p. 38, ^ .

San Cristoval: Wai-ai, Wainoni Bay. Ugi: Pawa. Bio. Malaita:
Auki. Ysabel: Fulakora.

An abundant species, nesting beneath stones, logs, or bark, or in
little shelters made of vegetable debris on leaves. It is fond of
attending scales. One nest sheltered two pupae of a myrmecophilous
fly, (Bardistopus sp. nov.) from which I reared adults.

Camponotinae.

93. AcROPYGA (Acropyga) moluccana Mayr, subsp.
papuana, subsp. nov.

Worker. Length 2.5-2.75 mm.

Differing from typical moluccana as figured by Emery (Ann. Mus. civ. stor.
nat. Genova, 1900, 40, p. 698' fig. 13a) only in the shape of the head, which
has the sides much less convex, more as in crassicornis Emery, but differing
from the latter in having occipital border concave and the penultimate anten-
nal joint a little longer than broad, instead of transverse. The tegument is
feebly shining. The color is brownish yellow.

Ugi: Pawa (Type-locality). Malaita: Auki. San Cristoval:
Wainoni Bay, Star Harbor. Ysabel: Fulakora.

A very common species, found in populous colonies beneath stones or
logs. Type— M. C. Z. 9,182.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 365

93a. AcROPYGA (Acropyga) moluccana Mayr, var. opaca Stitz.
Sitzungsb. Gesellsch. nat. freunde Berlin, 1911, p. 370, y .

Malaita: Interior near Gwountafu.

I am referring with doubt to this variety several workers and a
female which are darker in color and somewhat less shining than the
preceding form.

94. Acropyga (Rhizomyrma) lauta, sp. nov.

Worker. Length 1.50 mm.

Near oceanica Emery.

Head a little longer than broad, as broad in front as behind; sides feebly
convex; posterior angles rounded, border slightly concave. Mandibles nar-
row, with four separated, acute teeth. Clypeus truncate at middle of front,
obtusely angulate at side of truncated portion. Antennal scapes not attain-
ing occipital corners; first funicular joint more than twice as long as broad;
second joint campanulate, longer than broad; joints three to six a little broader
than long; terminal joint as long as the three preceding joints. Eyes minute,
situated in front of sides of head at anterior third. Thorax stout. Posterior
half of pronotum strongly convex, in front concave. Mesonotum longer than
broad, rather flat above. Epinotum with base convex and evenly rounding
into the declivity, which is flattened and twice as long as the base. Petiolar
node erect, narrow. Gaster elongate.

Shining throughout. Mandibles punctate. Head and gaster with dense,
fine, and shallow punctation, which is somewhat coarser on the gaster.

Head, gaster, and legs finely and densely pubescent, thorax and epinotum
sparsely so.

Color yellow; head slightly darker, mandibular teeth brownish to black.

Female. Length 2.25.

Head broader than long. Ocelli large. Frontal sulcus shallow. Eyes a
third as long as head, feebly convex. Antennal scapes slightly surpassing
occipital corners.

Wings (length 2.75 mm.) hyahne, veins and stigma fuscous.

San Cristoval : Pamua, Wainoni Bay. Malaita: Auki.
"Workers and females in a small series taken from beneath a stone
differ from Emery's descriptions and figure of oceanica in having the

366 bulletin: museum of comparative zoology.

head longer, the occipital border distinctly though shallowly concave
instead of truncate, and the frontal lobes divided by a broad triangular
spac^. Type.— ^l. C. Z. 9,183.

95. Plagiolepis longipes (Jerdon).
Formica longipes Jerdon, Madras journ. litt. sci., 1851, 17, p. 122, ^ .

Santa Cruz: Graciosa Bay. Throughout the Solomons.

Plagiolepis loncjipcs occurred in all the localities visited; it belongs
to the group of insects that is rapidly being distributed throughout the
world, and associated indirectly with man. Like the other members
of this anthropopliilus fauna, it prefers to nest in cleared land, under
stones, logs, or debris, or in the ground. I frequently found it in the
forest, but on every occasion, it is interesting to note, there were signs of
former native A'illages or clearings, indicating that the species invariably
establishes its colonies in the vicinity of human habitations.

96. Prenolepis (Nylanderia) longicornis (Latreille).

Formica longicornis Latr., Hist. nat. fourmis, 1802, p. 113, ^ .

Santa Cruz: Graciosa Bay. Santa Anna.
This species was found in all the localities visited.

97a. Prenolepis (Nylanderia) minutula Forel, subsp.

atomus Forel.

Mitt. Zool. mus. Berlin, 1901, 2, heft 1, p. 25, ^ .

Ugi: Pawa. Ysabel: Fulakora.
Occurs in small colonies beneath stones.

98. Prenolepis (Nylanderia) vividula Nylander.

Acta Soc. sci. Fenn., 1846, 2, p. 900, ^ 9 d".
Ugi : Pawa.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 367

99a. Prenolepis (Nylanderia) obscura Mayr, subsp.
BiSMARCKENSis Forel.

Mitt. Zool. mus. Berlin, 1901, 2, heft 1, p. 26, ^ 9 cf .

San Cristoval: Wainoni Bay. Ysabel: Fulakora.

100. Prenolepis (Nylanderia) stigmaticus, sp. nov.
IVorker. Length 2.25-3.25 mm.

Head elongate nearly as broad in front as behind; sides slightly convex;
posterior border nearly straight. Mandibles slender, arcuate with six acute
triangular teeth, the third and fifth distinctly smaller than the others. Gly-
peus convex, broadly rounded in front. Antennae very long and slender,

Fig. 41. — Prenolepis (Nylanderia) stigmaticus Mann. Worker. Lateral view of head,
thorax, and petiole.

their scapes surpassing occipital corners by five eighths their length; first
funicular joint nearly twice as long as'the second; joints three to seven about
four times as long as broad; joints eight to ten a little shorter; terminal joint
shorter than the two preceding joints together. Eyes large and convex, situ-
ated in front of sides a little anterior to middle. Pronotum about as long as
broad, sUghtly convex. Mesonotum much longer than broad; the flat meso-
epinotal suture strongly impressed. Basal portion of epinotum very convex
and rounding into the flat declivous surface which it equals in length; stigmata
situated at middle of sides of declivity prominently tuberculate. Petiolar
node broadly cuneiform.

Shining, with shallow microscopic punctures on head. Mandibles subtly
punctate.

368 bulletin: museum of comparative zoology.

Very coarse, stiff hairs moderately abundant on head, body, and appendages.
Color pale brownish yellow throughout.
Hairs black.

San Cristoval: Wai-ai (Type-locality). Ugi: Pawa. Three
Sisters : Malapaina.

A very distinct species, resembling steeli Forel in the variation in size
among workers from the same colony, but differing in having the anten-
nal joints longer, in the flattened mesonotum, and in sculpture.

101a. Oecophylla SMARAODiNAFabricius, var. subnitida Emery.
Ann. Soc. ent. France 1892, 60, p. 565, ^ .

Ugi: Pawa. San Cristoval: Wai-ai. Santa Anna. Three Sisters :
Malapaina. Bio. Malaita: Auki. Florida: Tulagi. Russell: Yan-
dina: West Bay. Ysabel: Fulakora. Guadalcanar: Rere.

This species was exceedingly abundant in certain localities. On
Malapaina, where trees were being felled, it occurred in such numbers
as to seriously interfere with collecting. Their nests had been
destroyed by the falling of the trees and the workers were very pug-
nacious. The larger workers are able to bite appreciably and, during
the several days that I spent collecting among the branches there
were few moments in which one or more were not biting me.

102. Opisthopsis manni Wheeler.
Bull. M. C. Z., 1918, 62, p. 361, pi. 3, fig. 25, 26, ^ 9 .

San Cristoval: W'ai-ai. Three Sisters: Malapaina. Malaita:
Auki. Russell: Yandina.

This is an arboreal species. The only colony that I found consisted
of about a dozen workers and a deiilated female. It was beneath a
piece of loose bark on a recently felled tree. The workers move with
such rapidity that it is almost impossible to collect them without
injury.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

369

103. Camponotus (Myrmoturba) maculatus Fabricius, subsp.
NOVA HOLLANDIAE Mayr, var. papua Emery.

Lorentz's Nova Guinea, 1911, 9, p. 256,01.

Santa Cruz: Graciosa Bay. San Cristoval: Wai-ai, Wainoni
Bay. Ysabel: Fulakora.

103a. Camponotus (Myrmoturba) maculatus Fabr., subsp.

SANCTAE CRUCIS, Subsp. nOV.

Soldier. Length 7.5-8 mm.

Head short and broad, slightly narrowed in front; broadly but very shal-
lowly excised posteriorly, posterior corners rather prominent. Mandibles
stout, 6-toothed. Cl>T)eus broadly rounded
at anterior border. Antennae slender, their
scapes extending about one fifth their length
past the occipital corners.

Shining; finely shagreened. Mandibles
shining, punctate, and striate apically. Cly-
peus, front, and cheeks with coarse piligerous
punctures.

Hairs yellow, erect, and sparse, long on
body, shorter on head; very sparse on
appendages.

Reddish brown; head and gaster a little
darker than the rest.

Santa Cruz
Anna.

Graciosa Bay, Santa ^"^- ^2 — Camponotus (Myrmo-
turba) maculatus sanctae crucis
Mann. Worker major. Front

Resembles 7iova hollandiae var. hinnilior view of head.
Forel, from Queensland, but has the

head broader and shorter, with the elypeal lobe less produced and
with its anterior border broadly rounded instead of truncate; the
head is not strongly infuscated as in humilior.

A very abundant ant, nesting in cavities of trees and in rotten wood.

104a. Camponotus (Mayrmamblys) reticulatus Roger, var,

BEDOTi Emery.

Camponotus bedoii Emery, Rev. Suisse zool., 1893, 1, p. 196, ^ .

370

bulletin: museum of comparative zoology.

Santa Cruz: Graciosa Bay. Santa Anna. Malaita: Auki. Russell:
Yandina. Guadalcanar: Rere. Bio. Florida: Tulagi. Rendova.
New Georgia. Rubiana Lagoon.

105. Camponotus (Colobopsis) guppyi, sp. nov.

Soldier. Length 5 mm.

Head longer than broad, narrowest in front, sides in front of eyes slightly
convex, truncated portion separated from the front by a rounded angle.
Mandibles 5-dentate. Clj^peus two thirds on truncated portion, margined

Fig. 43. — Camponotus (Colobopsis) guppyi Mann. Worker major,
head, thorax, and petiole.

Lateral view of

at sides; the posterior third longitudinally impressed at middle, carinate
between this impression and the lateral border; anterior portion with a sharp
median carina, anterior border straight. Frontal area broadly triangular.
Frontal carinae moderately divergent, extending to a point opposite middle
of eyes. Eyes situated on sides of front, back of posterior tliird of head.

Fig. 44. — Camponotus (Colobopsis) guppyi Mann. Worker minor. Lateral view ot
head, thorax, and petiole.

Antennae stout, their scapes surpassing occipital corners by a distance a little
greater than their diameter at tips. Pronotum rather flat, sides rounded.
Mesonotum a third broader than long. Mesoepinotal impression deep and
nearly half as long as the basal portion of epinotum. Base of epinotum
broader than long and truncate behind; in profile shorter than the concave

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 371

declivity, over which it sUghtly projects. Petiolar node in profile about as
long as high with dechvous anterior and slightly sloping posterior surface,
but little narrowed above; from above about twice as long as broad, with
narrowly rounded and submargined sides. Gaster elongate, oval. Legs
rather stout. • '

Moderately shining. Mandibles rugulosely striate. Anterior portions of
front and cheeks with stout rugose striae. Head subtly, reticulately striate,
the remainder with dense, though microscopic striae, which are seriolate on
thorax and transverse on gaster.

Head and gaster with a very few erect hairs.

Pale reddish brown, with fuscous blotches on the vertex, lateral blotches
on the gaster and dorsal ones on the base of the first, and the middle of the
second and margin of the third segments. Antennal scapes infuscated api-
cally; first funicular joint yellowish brown, remainder of funiculus black.
Femora dark reddish brown to black, except from a yellowish area on flexor
surface ; basal three eighths of tibiae yellow-brown the remainder dark brown,
almost black.

Worker (media). Length 4 mm.

Head a little longer than broad, slightly narrowed in front, with rounded
corners and posterior border. Clypeus convex, obscurely carinate at middle,
with nearly straight anterior border. Antennal scapes surpassing occipital
borders by nearly half their length. Pronotum a little broader than long.
Mesonotum longer than broad, separated from epinotum by a broad saddle-
shaped impression. Base of epinotum flat; at apex developed as a lamellate
projection, deeply concave and elevated and bilobed at tip (somewhat scoop-
shaped), in profile projecting over the very concave declivity about as far
as the declivity is high. Petiolar scale similar to that of soldier, but more
strongly narrowed at sides. Gaster elongate oval.

Less shining, than soldier, very densely and finely and seriolately striate
throughout.

Erect hairs very sparse on head and gaster.

Color as in soldier.

Worker (minor). Length 3 mm.

Differing from the above in its somewhat longer antennal scapes and in the
more extreme prolongation of the lamellate posterior border of the epinotum.

Malaita: Auki.

This extraordinary species is described from a small series of isolated
workers ^nd a single soldier, found on leaves at different times about
the government residence at Auki and on the trail to Aisisiki.

372 bulletin: museum of comparative zoology.

In the structure of the epinotum it is markedly different from all
other species belonging to the subgenus, and the coloration is exceed-
ingly characteristic. Type. — M. C. Z. 9,184.

106. Camponotus (Colobopsis) elysii, sp. nov.

Soldier. Length 6.5 mm.

Form rather slender.

Head a fourth longer than broad, sides subparallel, border slightly convex.
Mandibular blades with four teeth apically and two poorly defined ones
basally. Clypeus strongly carinate, somewhat narrowed and with an elevated
margin at anterior third of sides, anterior border nearly straight; posterior
three eighths above depressed portion barely more than twice as broad as long.
Antennal scapes nearly, but not quite attaining occipital corners. Frontal
area small, about twice as broad as long; frontal carinae slightly sinuous,
moderately divergent, extending to a point about opposite middle of eyes.
Eyes smaller than is usual in Colobopsis, flat, situated in front of sides, well
posterior to middle. Pronotmn and mesonotum moderately convex, each
broader than long. Thoracic sutures weakly impressed. Base of epinotum
broad and convex; declivity sUghtly concave, the two surfaces broadly round-
ing into each other. Petiolar node in profile, twice as high as long, narrowed,
submargined dorsally; from above, four times as broad as long. Gaster
elongate. Legs stout.

Shining. Mandibles finely rugulose and with sparse, large punctures.
Clypeus and anterior border of cheeks with short, rugose striae; remainder of
head and body subtly and very densely seriolately striolate, with scattered
distinct, though fine, punctures on front between carinae.

Pale yeUow-brown, the gaster and a spot on vertex darker.

Worker. Length 4 mm.

Head a little longer than broad, evenly rounded behind, cheeks parallel.
Clypeus convex, rounded anteriorly, not carinate. Antennal scapes sur-
passing occipital corners by a little less than half their length. Front strongly
impressed between the carinae. Frontal carinae short, nearly straight, ex-
tending to opposite anterior third of eyes. Base of epinotum compressed
and narrowly convex above, about four times as long as broad; in profile
narrowly rounded between base and declivity.

Petiolar node cuneiform, with convex anterior and straight posterior sur-
faces, dorsal margin acute.

Sculpture as in soldier, except that the rugae on clypeus and anterior por-
tion of cheeks are lacking.

Color pale reddish br^wn throughout.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 373

Three Sisters: Malapaina (Type-locality). San Cristoval: Wai-
noni Bay.

Near sommeri Forel, from New Caledonia and fidor Forel from
Australia, but with the head much narrower and more elongate.

The San Cristoval specimens are a little darker than those from
Malapaina, which may not be fully colored.

107. Camponotus (Colobopsis) loa, sp. nov.

Soldier. Length 4.5 mm.

Head longer than broad, sides subparallel, posterior angles broadly rounded,
border convex; rounded between anterior impressed portion and dorsal sur-
face. Mandibles 5-dentate. Clypeus with five eighths of its length on the
truncated portion, quadrangular, distinctly margined laterally and straight
at anterior border; strongly bicarinate for entire length, the carinae con-
verging anteriorly, and with two lateral carinae on basal three eighths.
Frontal area poorly defined, broadly triangular. Frontal carinae slightly
divergent, nearly straight, extending to a point a little in front of middle of
eyes. Antennal scapes surpassing occipital corners by a distance equal to
their width at tips; funicular joint as long as the second and third together.

Fig. 45. — Camponotus (Colobopsis) loa Mann. Worker major. Lateral view of head,
thorax, and petiole.

The remaining joints, except the terminal, gradually shorter and stouter;
terminal joint as long as the two preceding joints together. Eyes large and
flat; situated at posterior third of head. Thorax robust. Pronotura but
slightly convex above, humeri broadly rounded. Mesothorax rather flat,

374

bulletin: museum of comparative zoology.

one and a third times as broad as long. Base of epinotum moderately convex,
about as long as broad, rounding into the declivity, which is flat and a little
longer than base. Petiolar scale in profile less than twice as high as thick,
slightly convex in front, straight behind and flattened dorsally; from above,
twice as broad as long, narrowly rounded and submargined laterally. Gaster
elongate, legs stout.

Moderately shining. Mandibles rugulose and striolate. Truncated por-
tion of head more shining than the rest, the impressed surfaces lateral to the
clypeus uneven and shallowly punctate. Cheeks anteriorly with a series of
short, irregular rugose carinae; the remainder of head shallowly and reticu-
lately punctate. Thorax and abdomen very subtly reticulostriolate.

Head and body with a very few long erect hairs. Fine and short recumbent
hairs on head and gaster.

Fig. 46. — Camponotus (Colobopsis) loa Mann. Worker minor. Lateral view of head,
thorax, and petiole.

Reddish brown; mandibular teeth black; head with a fuscous blotch on
apex; basal two thirds of first, and apical two thirds of second and apical
one half of remaining gastric segments black.

Worker {minor). Length 2.5-3 mm.

Head longer than broad, slightly narrowest in front, occipital corners
rounded, border nearly straight. Clj^eus large and convex, very obtusely
carinate, anterior border shallowly concave at middle. Frontal carinae
moderately divergent, extending to a point opposite anterior third of eyes.
Antennal scapes sm'passing occipital corners by about one third their length.
Eyes large, Uttle convex, situated in front of sides a little posterior to middle.
Pronotum rather flat above, broadest at middle with sides narrowly rounded.
Mesonotum broader than long, moderately narrowed behind, with straight
sides. Base of epinotum compressed and narrowly convex; produced at apex
as a short slightly projecting connvile; declivity concave in profile and a little
shorter than the base. Petiole and gaster as in soldier.

Moderately shining. Gaster with very dense, short transverse striolae,
the remainder of body and the head subtly and seriolately striolate.

Sparse, scattered, erect hairs, and very fine and short recumbent hairs on
head and gaster.

Pale yellow-brown; head posterior to antennal insertions darker, gaster
black, each segment with a yellowish band apically.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 375

San Cristoval: Wai-ai (Type-locality). Santa Cruz: Graciosa
Bay. Santa Anna. Three Sisters: Malapaina. Ugi: Pawa. Mal-
aita: Auki.

Described from numerous soldiers and workers taken from colonies
nesting in hollow vines at Wai-ai and from workers and soldiers found
running on leaves. There is considerable variation in the amount of
infuscation on the head and the Malapaina and Auki specimens are
pale in color, perhaps being varietally different, but more probably
not fully colored.

Cavrponotus loci is near C. conithorax Emery, from the New Hebrides,
but differs in size, color, and in the more convex base of epinotum.
Ca7nponotus conica Mayr. from Tonga has the epinotum similarly
shaped in the worker but the petiolar node is armed with two spines.
Type.— M. C. Z. 9,185.

107a. Camponotus (Colobopsis) loa Mann, var. belli, var. nov.

Several workers and soldiers found running about on the trunks of a recently
felled tree agree in habits and sculpture with the type form, but in the soldier
the front of head, the cheeks, thorax, epinotxim, and petiole, and the middle
and hind femora are dark brown, almost black. The anterior femora are
mostly black, but blotched with yellowish brown. The light-colored bands
on the otherwise black gaster are much narrower. The worker is black,
except on the extreme anterior border of cljrpeus, the mandibles, antennae,
tibiae, and tarsi, which are pale brown.

Florida: Tulagi.

108. PoLYRHACHis (Hedomyrma) santschi, sp. nov.

Worker. Length 5 mm.

Head longer than broad, broadly rounded behind, sides in front of eyes
moderately convex. Clypeus subcarinate at middle, broadly sloping at sides,
bidentate at middle of anterior border. Antennal scapes surpassing occipital
corners by more than half their length. Pronotum longer than broad, slightly
convex in profile, margined laterally; humeral spines broadened basally with
the inner corners angulate, apical three fourths slender, compressed and acute
at tips; extending forward and outward and bent downward. Mesonotum
and basal portion of epinotmn flat in front, shallowly concave between the
spines, together shorter than the declivity; spiracular tubercles at margin of

376 bulletin: museum of comparative zoology.

declivity large. Epinotal spines long, stout at basal two fifths, then slender;
little divergent, their distance apart at tips about equal to their length, directed
backward and upward. Petiole broader than long, sloping above, anterior
border straight, posterior border curved and at middle subgibbous; spines
rather thick at basal half, slender apically, a third longer than their distance
apart at base, extended parallel to sides of first gastric segment.

Moderately shining. Mandibles finely striate. Head, thorax, and base
of epinotum with strong, regular and entire striae, which are transverse on
posterior portion on cheeks, oblique on clypeus, and longitudinal on thorax
and epinotum. Epinotal declivity and posterior surface of petiole rugose.
Gaster finely and densely rugulose.

Front and vertex with very sparse, long, erect hairs. Head and thorax
with a thick covering of rich golden pubescence.

Black throughout.

Florida: Maliali.

Described from a unique worker. The curious, strong, and regular
sculpture and the slender downward, curved humeral spines are very
characteristic.

108a. PoLYRHACHis (Hedomyrma) santschi Mann, subsp. camp-

BELLi, subsp. nov.

Several workers differ from the typical form in having the golden pubescence
on gaster more dense, the petiolar spines shorter and in the sculpture of the
head. The striae of vertex extend longitudinally back to the posterior border
of the occiput at the middle where they are met obUquely by transverse striae
that extend down the sides of the occiput and cheeks. In santschi the occipital
region is transversely striate.

Russell: West Bay.

In both forms of the species the epinotal spines are strongly longi-
tudinally striate on the basal half. Type. — M. C. Z. 9,192.

109. PoLYRHACHis (Hedomyrma) geminatus, sp. nov.

Worker. Length 5 mm.

Head slightly longer than broad, sides in front of eyes feebly convex, poste-
rior border broadly rounded. Clypeus convex, obtusely carinate; lobed and
acutely bidentate at middle of anterior border. Frontal carinae not widely
separated, moderately elevated, parallel behind. Eyes convex, situated at

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

377

posterior corners of head. Pronotum narrowly margined at sides, one and
one third times broader than long, broadest behind; sides convex at posterior
two thirds, in front concave; humeri developed as broad, lamellate lobes with
elevated margins, apically concave and bidentate,
the outer tooth elongate and spinose, the inner
one rounded at tip. Mesoepinotum without
suture or constriction, much shorter than pro-
notum, flat except between spines where it is
concave. Basal and declivous portions of epino-
tum rounding into each other; declivity much
longer than base. Epinotal spines less than
twice as long as their distance apart at base,
stout, rather strongly curved downward. Petiolar
node in profile with anterior and posterior sur-
faces subparallel; dorsal surface slightly convex,
less than twice as broad as long, separated from
posterior surface by an obtuse margin; spines a
little shorter than their distance apart at base,
stout, arcuate.

Sublucid. Mandibles finely striate. Clypeus
with fine oblique striae. Head, thorax, and
epinotum with regular, slightly rugose striae,
which are longitudinal except on sides of prono-
tum where they are oblique. Petiolar node finely
striate. Gaster densely and rather coarsely and
rugosely striate longitudinally. Legs and antennae finely punctate.

Erect hairs coarse, very sparse on head and apical portion of gaster.
and body, with sparse, recumbent pubescence.

Black throughout.

Fig. 47. — Polyrhachis (Hedo-
myrma) geminatus Mann.
Worker. Thorax and

petiole from above.

Head

Florida: Maliali.
Described from a unique worker.

The species is distinguished by the short, upturned, and bifurcate
humeral spines, and the strong sculpture of the gaster.

110. Polyrhachis (Hedomyrma) annae, sp. nov.

Worker. Length 4.5-5 mm.

Head distinctly longer than broad, evenly rounded behind, sides in front
of eyes nearly straight, somewhat convergent. Clypeus evenly and moderately
convex, the anterior border bidentate and shallowly concave at middle.
Frontal carinae only slighty divergent, extending to opposite middle of eyes.
Scapes surpassing occipital border by more than two thirds of their length.

378

bulletin: museum of comparative zoology.

Pronotum broader than long, broadest behind middle, margined at sides;

hmneral spines flat, acute at tips, divergent, and pointing slightly upward.

Mesoepinotum broadest in front, sides margined, converging to base of spines,

surface between spines shaUowly concave, in front nearly flat. Spines flat-
tened and margined basally, slender api-
cally and acute at tips, a little longer than
their distance apart at base, moderately
divergent and extending slightly upward.
Petiole nearly flat, about the anterior
border truncate; posterior border feebly
rounded and slightly elevated and sub-
gibbous at middle; spines slender, parallel
with sides of gaster, slightly longer than
their distance apart at base.

Moderately shining. Mandibles densely
striate.

Head and pronotum densely rugulose-
striate the latter more coarsely and with
the striae more regular. Mesoepinotum
broadly and shallowly and reticulately
striate, petiole and first gastric segment
densely punctate.

Front and vertex with paired, erect setae.
Head and pronotum with rather thin
pubescence and mesoepinotum, petiole,
and first gastric segment with thicker
very fine pubescence, silvery except on
gaster where it is golden.

Black; appendages dark brownish red.

Fig. 48. — Polyrhachis (Hedomyr-
ma) annae Mann. Worker.
Thorax and petiole from above.

Santa Cruz: Graciosa Bay (Type-locality). San Cristoval:
Pamua, Wainoni Bay. Three Sisters: Malapaina.

Described from workers in a colony that rested in earth among the
suckers of an epiphyte and others taken on leaves and tree-trunks.
Type.— M. C. Z. 9,193.

This species is common and widely distributed, but the colonies
nest usually high in trees and are difficult to find.

111. Polyrhachis (Myrmothrinax) dahli Forel.
Mitt. Zool. mus. Beriin, 1901, 2, heft'l, p. 30, ^ .

San Cristoval: Wainoni Bay. Three Sisters: Malapaina. Mal-
aita: Auki. Florida: Tulagi.

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS.

379

Rare, though evidently of wide distribution. The workers in a small
series before me are a little smaller (length 6 mm.), but otherwise
agree closely with Forel's description, based on Bismarck Archipelago
specimens.

112. PoLYRHACHis (Myrma) salomo Forcl.
Rev. Suisse zooL, 1910, 18, p. 87, ^ .

Malaita: Auki. Russell: West Bay, Yandina.

Fig. 49. — Polyrhachis (Myrma) salomo Forel. Worker. Front view of head.

Nests usually in crevices of logs of trees, and builds partitions and
protecting walls of silk interwpven with vegetable material. One
colony was found beneath a stone.

113. Polyrhachis (Myrma) similis Viehmeyer.

Abh. berichte K. zool. anthr.-ethn. mus. Dresden, 1912, 14, no. 1, p. 8, ^ .

San Cristoval: Wai-ai, Pamua, Wainoni Bay, Star Harbor. Bio.
Ugi: Pawa. Florida: Tulagi, Maliali. Ysabel: Fulakora. Russell:
West Bay, Yandina. New Georgia. Rubiana Lagoon.

380 bulletin: museum of comparative zoology.

Probably the most conspicuous ant in the islands. Foraging workers
were on nearly every tree examined. The colonies are in tree-cavities.

114a. PoLYRHACHis (Myrma) relucens Latreille, subsp. Andro-
mache F. Smith, var. nesiotis, var. nov.

Worker.

Differing from subsp. andromache in having the legs darker in color, dark
red, with the femora in part and the tibiae infuscated and the tarsi black.

Santa Cruz : Graciosa Bay.

This variety, the only one of the relucens group found at Graciosa
Bay is exceedingly abundant and very large series were obtained, all
of which are constant in coloration. It nests in crevices of trees and
in hollow fern-stalks. Type — M. C. Z. 9,194.

Fig. 50. — Polyrhachis (Myrma) relucens andromache nesiotis Mann. Worker.

114b. Polyrhachis (Myrma) relucens Latreille, subsp. litigiosa

Emery.

Ann. Mus. civ. stor. nat. Geneva, 1897, 38, p. 581, S .

Ysabel: Fulakora.

Numerous colonies were found nesting in rotten logs. Some of the
cavities used as brood chambers were lined with silk.

Pinned specimens are very liable to grease and many among my
series have the gaster black in appearance due to this, which hides the
rich golden pubescence characteristic of the subspecies.

•MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 381

115. PoLYRHACHis (Charyomyrma) rere, sp. nov.

Worker. Length 3.5-4 mm.

Head a little longer than broad and somewhat narrowed in front, sides in
front of eyes feebly convex, posterior border broadly rounded. Clypeus cari-
nate, front lobed, the middle portion concave and obtusely dentate at sides.
Front carinate at middle between frontal carinae. Frontal carinae widely
separated, moderately elevated and little divergent behind. Eyes strongly

Fig. 51. — Polyrhachis (Charyomyrma) rere Mann. Worker. Lateral view of thorax
and petiole.

convex, situated at posterior corners of head. Pronotum twice as broad as
long, slightly convex, narrowly margined at sides ; humeri with small triangular
teeth that are a little longer than broad. Meso- and epinotum without suture,
flat above, narrowly margined at sides. Base of epinotum very broadly
rounding into the dechvity. Epinotal spines a little longer than half their
distance apart at base, slender, divergent, and nearly straight. Petiolar node
mth flat anterior and posterior surfaces, strongly narrowed and submargined
above; seen from the front, the border is obtusely angulate at middle; spines
short, rather stout, arcuate, and parallel with base of gaster.

Opaque, except gaster which is moderately shining. Mandibles with coarse,
separated striae and, at apex, large punctures. Head densely striolate-
rugose. Thorax and epinotum similarly but not as coarsely sculptured. Epi-
notal and petiolar spines and dorsal margin of petiole shining and rugose.
Gaster, legs, and antennae densely punctate.

Fine, long, erect pile, abundant on head, body, and appendage, and rather
long, appressed silvery pubescence on head and body, thick on gaster and
sparser elsewhere.

Black throughout.

Guadalcanar : Rere.

382

bulletin: museum of comparative zoology.

116. PoLYRHACHis (Charyomyrma) kaipi, sp. nov.

Worker. Length 5 mm.

Head a little longer than broad, sides convex, posterior border broadly
rounded. Mandibles with four large teeth. Clypeus carinate; anterior
border lobed and concave at middle. Frontal carinae short, the anterior two
thirds strongly laminate. Eyes moderately conve.x; situated at posterior
thirds of sides. Pronotum twice as broad as long; disc feebly convex; sides
and anterior border, except at middle, broadly lamellate, and moderately ele-
vated, humeral spines triangular and acute. Promesonotal suture narrowly

impressed. Mesonotum broader than long, broad-
est in front, sides lamellate and elevated similar
to but not as broadly as on pronotum. Base of
epinotum lower than mesonotum and distinct from
it, though the suture is feebly impressed; surface
fiat, twice as broad as long, acutely margined at
sides; broadly rounding into the flattened decliv-
ous portion which, in profile, is three times as long
as base; spines about as long as their distance
from base to promesonotal suture; thick basaUy
and becoming slender apically, slightly curved
downward, directed backward and moderately
divergent. Petiole narrowly rounded above, the
spine slender, about as long as those of epinotum
and encircling base of gaster. Gaster short and
ovate.

Strongly shining. Mandibles with regular, dense
rather coarse striae. Clypeus with irregular,
broad punctures, some of them confluent. Head
striate, the striae strong and longitudinal on
sides and cheeks, transversely arcuate above and
obsolete on the median portion of vertex and
occiput; striae of front finer and very irregular;
vertex with coarse, piligerous punctures. Thorax and abdomen not striate,
but with widely separated, foveolate punctures, similar to those on vertex.

Epinotum and posterior surface of petiole with a thin mat of long, silky,
semirecumbent hairs, and long, erect pile, the remainder with erect, very long,
fine, and rather abundant pilosity, shortest on the head and appendages, but
lacking the mat of shorter semirecumbent hairs.
Black; appendages dark reddish brown.
Pilosity yellowish.

Fig. 62. — Polyrhachis (Char-
yomyrma) kaipi Mann.
Worker. Thorax and
petiole from above.

Malaita: Auki.

MANN : ANTS OF THE BRITISH SOLOMON ISLANDS. 383

Described from a unique worker.

This species is near P. scutulatus F. Smith, from Aru. It agrees
with Smith's description except that the epinotum of kaipi is not
opaque, but shining. Emery has described and figured (Rev. Suisse
zool., 1893, 1, p. 226, fig. 7-8) a New Guinea specimen that he con-
siders scutulatus. In his figure of the head, the frontal carinae are
broadly curved at sides, the thorax is distinctly broader than the
head; the mesonotum is not distinctly separated from the epinotum
and is more than twice as broad as long; in kaipi the frontal cai'inae
are narrowly curved, the thorax is barely broader than the head and
the mesonotum is less than twice as broad as long. The epinotal
and petiolar spines of kaipi, as well as the humeral spines are very
much longer than in scutulatus. Poli/rhachis aurita Emery, also closely
related, has the thorax proportionally very much broader, and is
diff"erently sculptured with the epinotal declivity and posterior surface
of petiole striate, instead of smooth as in kaipi.

117. POLYRHACHIS (MyRMATOPA) ULYSSES Forcl.
Rev. Suisse zool., 1910, 18, p. 91, ^ .

Female. Length 10 mm.

Ocelli very small. Mesonotum convexly declivous at anterior third;
posterior two thirds slightly convex and narrowly and very feebly margined
at sides. Scutellum transversely oval. Prescutellum with weakly margined

Fig. 53. — Polyrhachis (Myrmatopa) ulysses ForeL Petiole from front.

anterior border. Base of epinotum thinly margined at sides, the margins
lacking the anterior angulate structure found in the worker, spines more
slender than in the worker. Petiole similar to that of worker. Wings (length
9.5 mm.) clear, veins and pterostigma brown.

384

bulletin: museum of comparative zoology.

Florida: Tulagi, Maliali.

A handsome species, conspicuous because of the white tips to the
antennae. It is much less common than the following related species.

118. PoLYRHACHis (Myrmatopa) osae, sp. nov.

Worker. Length 6.75-7.5 mm.

Head in structure very similar to that of ulysses. Pronotum convex; humeri
with indistinct denticulae. Mesonotum margined at sides, the margins
elevated into broad, subtriangular lobes; surface concave between lobes,
behind declivous to the mesoepinotal suture. Epinotum bluntly margined

Fig. 54. — Polyrhachis (Myrmatopa) osae Mann. Worker.

at sides ; margins somewhat elevated, but not angulate in
front, spines short and triangular; declivity sloping, much
longer than base, with rather flat surface. Petiole as in
campbelH but the spines shorter and less divergent (their
distance at tips about equal to their length).

Gaster very shining, the rest moderately shining except
the head, which is nearly opaque. Mandibles punctate
and striolate. Clypeus shallowly, cribrately punctate, ex-
cept for a space near the base. Head rugulosely punctate.
Thorax, epinotum, and petiole shallowly punctate and
reticulate, the petiole more coarsely than the rest. An-
tennae and legs finely striolate, punctate.

Short, erect hairs present on funiculus and very fine
and short appressed ones on clypeus and gaster and a pair
of erect setae on sides of basal portion of clypeus.

Black. Mandibles and legs obscurely rufous last six
joints of antennae pale yellow-brown.

Feviale.

Length 8 mm.

Fig. 55. — Polyrha-
chis (Myrmatopa)

osae Mann. Head as m worker. Ocelli small. Mesothorax slightly

Worker. convex abovc) rounded at sides, with distinct parapsidal

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 385

furrows. Scutellum a little broader than long, with nearly straight anterior
border. Basal portion of epinotum roundly margined at apical half; spines
as in worker. Petiole similar to that of worker.

Feebly shining, with sculpture as in worker.

Color as in worker. Wings weakly infuscated; veins and pterostigma
brown.

Ugi: Pawa (Type-locality). San Cristoval: Wai-ai, Pamua,
Wainoni Bay.

Closely related to P. ulysses Forel, but distinct in epinotal structure
and in the form of the petiolar spines, which are not as long and are
very much less divergent than in that species. The thorax is also
much more shining in osae than in ulysses. Type. — M. C. Z. 9,191.

Fig. 56. — Polyrhachis (Myrmatopa) osae Mann. Worker. Petiole from front.

This species, which is named for Mrs. Osa Martin-Johnson, the
first American woman to travel in the Solomons, is a characteristic
species of the extreme eastern end of the group. The nests, built on
leaves, usually on palm trees, are of carton, strengthened and lined,
as well as fastened to the leaf by silk, and are very common objects
in the forest. If a leaf bearing a nest is broken off and laid on the
ground the ants rush out, but soon enter the nest again and will remain
there, even though the nest be placed in a coat pocket and carried.
I examined large numbers of nests of this, as well as other species of
Polyrhachis, in search of inquilines, wnthout finding any.

On several occasions I found two dealated females of osae, together
with larvae, in very small nests, which indicates that the colonies
are sometimes started jointly by more than one queen.

Nests are shown on Plates 1, 2.

386 bulletin: museum of comparative zoology.

119. POLYRHACHIS (DoLICHORHACHIS, Subgen. nOV.) MALAENSIS,

sp. nov.

Worker. Length 6.5-7.5 mm.

Head oval, longer than broad, rather narrowly rounded behind. Cheeks
evenly convex. Mandibles 5-dentate. Clypeus distinctly carinate at middle;
anterior border notched at middle and angulately toothed at either side of
notch. Frontal carina narrow, moderately elevated, not divergent, extending
to opposite middle of eyes. Antennae long and slender. Eyes small, convex,
situated in front of side at posterior fourth of head. Thorax slender, flat
above, sides strongly, angulately margined, but not laminate. Pronotum
longer than broad, humeri with slender curved spines as long as their distance
from base to promesonotal suture, directed forward and upward and moder-
ately divergent. Promesonotal suture very distinctly, though narrowly

Fig. 57.^ Polyrhachis (Dolichorhachis) malaeasis Mann. Worker. Lateral view of
thorax and petiole.

impressed. Mesonotum longer than broad, narrowed behind; sides in front
convex, behind concave. Mesoepinotal suture indicated by a distinct trans-
verse carina. Base of epinotum quadrate, longer than broad, with straight,
sharply margined sides, concave between spines and evenly rounding into the
declivity; declivity with convex surface, about as long as base. Spines long
and very slender, extending backward and slightly upward and but little
divergent. Petiolar node thick, anterior face in profile straight and forming
a prominent blunt angle with the flat dorsal surface; posterior surface convex;
spines very slender, longer than those of epinotum, moderately divergent,
extending backward and upward and curved inward. Gaster broadly oval.
Legs slender.

Subopaque. Mandibles densely rugulose and striate apically. Clypeus
and head rugulose striate, the striae dense, oblique on clypeus and longitudinal
on head. Thorax, epinotum, and petiole striate similar to head. Gaster

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 387

very finely and densely punctate and striate. Appendages finely and densely
punctate.

Antennae and tibiae pollinose, head and body clothed densely with fine,
short, silky, glistening silver pubescence.

Color black throughout.

Female (dealated). Length 9 mm.

Ocelli very small. Thorax and epinotum slender, but much shorter than
in the worker. Otherwise similar to worker with the usual sexual modifica-
tions.

Malaita: Interior near Gwountafu.

Described from two workers and a female found in the ground near
our camp fire, at an altitude of about 2,200 feet.

The elongate form and the arrangement of the spines are similar to
P. pJiyllophila F. Smith and related Oriental species included in
the subgenus Mynnhopla. However the margined thorax and epi-
notum do not permit including it in this subgenus and make the erec-
tion of Dolichorhachis necessary.

120. PoLYRHACHis (Myrmhopla) wheeleri, sp. nov.

Worker. Length 6.-6.5 mm.

Head a little longer than broad, broadly rounded behind eyes; sides in front
of eyes slightly convex. Mandibles 5-dentate. Clypeus obtusely, though
distinctly carinate, with an anterior median lobe, separated from the rest by

Fig. 58. — Polyrhachis (Myrmhopla) wheeleri Mann. Worker,

a moderately impressed transverse suture, and strongly bidentate. Frontal
area indistinctly defined. Frontal carinae, short, extending to opposite
middle of eyes, strongly and angulately elevated behind antennal insertions.
Eyes strongly convex, situated at posterior third of sides. Antennae long,
thorax robust. Pronotum slightly convex above, not margined, with long,

388 bulletin: m.useum of compakative zoology.

rather stout straight spines extending forward, upward, and outward. Pro-
mesonotal suture narrow. Mesoepinotal impression very feeble. Base of
epinotum twice as broad as long, slightly convex between the spines which are
stout, about one and one half times as long as pronotal spines, moderately
divergent, extending backward and upward and very feebly curved at apical
half. Petiolar node in profile evenly convex in front, nearly straight behind;
spines stout, strongly curved upward and backward, but not sufficiently di-
vergent to encircle base of gaster, with two stout, triangular acuminate teeth
between. Gaster very broadly oval.

Mandibles moderately shining, striolate, with elongate punctures; apically,
with several fovolate punctures. Head, thorax, epinotum, and petiole opaque,
coarsely and densely reticulately rugose; spines strongly striate and rugose.
Gaster shining, finely and regularly punctate. Antennal scapes densely
punctate. Legs transversely striolate.

Funiculus poUinose. Gaster with sparse, very fine, and short recumbent
hairs. Other hairs absent except a few setae on front of head and clypeus.

Black; legs very dark ferruginous.

Malaita: Near Auki, on trail to x\i.slsiki. Ysabel; Fulakora.

One large colony was found in each locality. The Malaita nest v^^as
built on the under side of an agave leaf, mostly of silk but with a few
strand? of vegetable material through it, and resembled an inverted
tent. The colony at Fulakora was in a triangular nest made of two
leaves connected by a sheet of silk. Partitions of pure silk divided
the interior into three chambers. Type. — M. C. Z. 9,188.

When I shook the bush on which this was situated the workers
rushed out and grouped themselves on the top of the nest, standing
with the thorax elevated, and the gaster shoved forward, and kept
the antennae and the forelegs waving. They made an appreciable
amount of noise when they rushed about, the nest serving as a sounding
box.

This species resembles armata F. Smith in habitus but is a much
smaller form, with shorter pronotal spines, finer thoracic sculpture and
very shining gaster.

121. PoLYKHACHis (Myrmhopla) argentea Mayr.

Verh. K.K. zool.-bot. gesellsch. Wien, 1862, 12, p. 682, g .

Malaita: Auki. Florida: Mahali.

I found argentea only on two occasions. A small colony at Auki
was in a nest made of two leaves fastened together with silk, similar

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 389

to the one described by Wroughton (Journ. Bombay nat. hist, soc,
1892. 7, p. 37).

My specimens agree perfectly with Mayr's description and with
workers from the Phihppines.

122a. PoLYRHACHis (Cyrtomyrma) rastellata (Latreille), var.
FULAKORA, var. nov.

Worker. Length 6.5 mm.

A very large series of workers from many colonies agree with rastellata in
structure, having the pronotum broadest in front and with rounded humeri,
the epinotum entirely without spines and the lateral spines of the petiole well-
developed and acute, but differ in the color of the legs, which are very dark
reddish brown, instead of blood-red. This difference is constant throughout
the series and warrants this variety.

Ysabel : Fulakora.

Stitz (Sitzungsb. Gesellsch. nat. freunde Beriin, 1911, p. 381) describes
rastellata var. major as larger than the typical form and with dark
brown legs. He does not give the length. The humeri of his variety
are more angular than in rastellata. In var. fulakora they are rounded
as in the type form. The epinotum in fulakora is more convex in profile
rather similar to subsp. euryala F. Smith from Java as figured by Vieh-
meyer (Archiv naturg., 19i3, p. 52). Tijpe.— M. C. Z. 9,189.

122b. PoLYRHACHis (Cyrtomyrma) RASTELLATA Latreille, subsp.

UGiENSis, subsp. nov.

Worker. Length 5.5-6 mm.

The specimens from the eastern Solomons represent still another form of
this variable species, characterized by the shorter spines of the petiole, which
in some specimens have the lateral ones reduced to mere angles, and in the
color of the legs. The femora and the tibiae are reddish brown, with the bases
of the latter and the tarsi black. The antennae are black, except the extreme
tip of the last segment, which is brown.

Nests are figured on Plate 2, fig. 1, 2.

Ugi : Pawa. San Cristoval : Wai-ai, Parnua, Wainoni Bay. Three
Sisters: Malapaina.

390 bulletin: museum of comparative zoology.

The different forms of rastellata build nests of vegetable fibres inter-
woven with silk, selecting as a site the tip of a leaf, usually that of a
palm. The sides of the leaf are drawn somewhat together and fast-
ened with silk form a partial and sometimes an entire covering of the
nest. Type — M. C. Z. 9,190.

122c. POLYRHACHIS (CyRATOMYRMa) RASTELLATA Latreille, Var. JOHN-

soNi, var. nov.
Worker. Length 7 mm.

Resembling laevior var. debilis Emery in having on the epinotum a pair of
distinct though very short and small spines. It is much larger in size than
debilis and the legs are uniformly dark brown, almost black.

The thorax is evenly arched, much as in leonidas Forel and the epinotal
declivity in profile is straight.

Rendova.

One colony was found, in a silk and carton nest on a palm leaf.
Type — M.C.Z. 9,195.

123. PoLYRHACHis (Cyrtomyrma) emeryana, sp. nov.

Worker. Length 5 mm.

Head a little broader than long, broadly rounded behind and rather strongly
narrowed in front. Cl3rpeus only moderately convex, shallowly concave at
middle of anterior border. Frontal carinae little divergent, extending to

Fig. 59. — Polyrhachis (Cyrtomyrma) emeryana Mann. Worker. Lateral view of
thorax and petiole.

opposite middle of eyes, in profile elevated and evenly rounded in front.
Antennal scapes surpassing occipital corners by about five eighths of their
length. Pronotmn broader than long, strongly convex above. Mesoepinotum

MANN: ANTS OF THE BRITISH SOLOMON ISLANDS. 391

little convex, the suture indicated by a very shallow and indistinct impression;
subangulate between base and declivity. Epinotal spines nearly straight, a
little shorter than their distance apart at base. Petiolar node subtriangular
in profile, the dorsal edge margined and bearing four spines, the inner pair of
which are triangular, as broad basally as long, and the outer elongate-triangu-
lar, acute apically and four times as long as the inner pair.

Moderately shining. Finely coriaceous and with minute punctation.

Very fine and short recumbent pubescence thinly distributed on head, body,
and appendages.

Color black.

Malaita: Auki.

Described from a unique worker.

This species is near leonidas Forel, but is distinctly smaller, the meso-
epinotum is much less convex, the spines smaller and more cylindrical
and the outer pair of petiolar spines very much longer in proportion
to the inner pair than in a cotype of leonidas in Professor Wheeler's
collection.

EXPLANATION OF THE PLATES.

PLATE 1.

Mann. — Ants of the British Solomons.

PLATE 1.

Fig. 1-3. — Nests of Polyrhachis (Myrrnatopa) osae Mann. 1^ X nat. size.
Photographs by John Howard Paine.

LU
1-
<

(/)

I-

J<5

:0
O

o
o

3
QQ

PLATE 2.

Mann. — Ants of the British Solomons.

PLATE 2.

Fig. 1. — Nest of Polyrhachis {C yrtomyrma) rastellata (Latreille), subsp. ugiensis
Mann on leaf of a palm. 1| X nat. size.

Fig. 2. — Incipient nest of the same. 1| X nat. size.

Fig. 3. — Incipient nest of Polyrhachis. {Myrmato-pa) osae Mann. IJ X nat.
size.

Photographs by John Howard Paine.

BULL. MUS. COMP. ZOOL.

Mann. Ants. Plate 2

■>?5vl

'Ji

:f:

)r.

M I

^.r^'

JUL 211

HARVARD

JNIVERSITY
Bulletin of the Museum of Comparative Zoology ^ I • L»

AT HARVARD COLLEGE.
Vol. LXIII. No. 8.

SOME AMPHIBIANS FROM NORTHWESTERN PERU, WITH
A REVISION OF THE GENERA PHYLLOBATES AND

TELMATOBIUS.

By Thomas Barbour and G. K. Noble.

With Thbee Plates.

CAMBRIDGE, MASS., U. S. A.

PRINTED FOR THE MUSEUM.

January, 1920.

No. 8. — Sovie Amphibians from Northwestern Peru, with a Revision
of the Genera PhyUobates and Telmatobius.

By Thomas Barbour and G. K. Noble.

During the summer and autumn of 1916 the junior author served
as zoologist of an expedition to northwestern Peru undertaken in
the interests of the School of Tropical Medicine (Harvard University)
and the Museum.

This paper is the first of a series dealing with the herpetological
collections secured. It is our intention to make these papers more
than faunal lists and though it was expected that the deserts of
northern Peru would yield few amphibians and that the number of
species would be small, the percentage of new forms proves extra-
ordinarily large. Notes on the habits of the species observed, especi-
ally of the new marsupial frogs will be included in a later paper.

The expedition crossed the provinces of Piura, Cajamarca, and
Lambayeque. The towns of Huancabamba and Palambla are on
the western range of the Andes, on the border of Piura. This north-
ern Huancabamba should not be confused with the town of the same
name of central Peru near Oxapampa. From the latter several reptiles
and amphibians were collected by Enrique Boettger in 1910 and
described by Boulenger. Not one of these species was found in the
Huancabamba visited. This caused some confusion and after cor-
respondence with Dr. Boulenger and Mr. W. F. H. Rosenberg, it is
apparent that Boettger's material should be labeled Oxapampa, his
Huancabamba being far less widely known than the much larger
town of the same name. The species affected are : — Anolis boettgeri,
Stenoeerciis boettgeri, Prionodactylus sjyinalis (Ann. mag. nat. hist.,
1911, ser. 8, 7, p. 19-24); Hyla melanopleura, Edalorhina nasuta
{Loc. cit., 1912, ser. 8, 12, p. 185-190); Leptognathus polylepis, Lachesis
chloromelas (Loc. cit., 1912, ser. 8, 10, p. 422-424); and IlyJella occllata
(Loc. cit., 1918, ser. 9, 2, p. 433). Tabaconas lies in a little valley
between the ranges of the Cordillera in the northern part of Cajamarca
and is in the only strip of rain forest met with during the expedition.
Perico, Bellavista, and Chumayo are in the same province in the low,
broad and arid valleys of the Chinchipe and Maranon Rivers. Quero-

396 bulletin: museum of comparative zoology.

'cotilla is on the border line between Cajamarea and Lambayeque.
Like Huaneabamba it is a tiiountain town on the western range of the
Andes.

In the preparation of this paper we are greatly indebted, especially
in our study of the genera Phyllobates and Tehnatobius, to Dr.
Lettnhard Stejneger for the loan of a number of specimens; to Mr.
H. W. Fowler for a similar loan of Cope's types of Tdmatohiiis angus-
tipcs, T. pustulosus, and T. acmaricus; to Dr. G. A. Boulenger for his
unpublished notes on Phi/Ilobates pratti and to Dr. Louis Roule for a
sketch of the pectoral girdle of P. bicolor.

Ranidae.
Phyllobates sylvatica, sp. nov.

Diagnosis. Closely related to P. latinasus (Cope) from which it may be
distinguished by the following characters : —

Tj^mpanum about one third instead of one half the diameter of the eye.
First finger a trifle shorter instead of a little longer than the second. Tibio-
tarsal articulation reaching to the middle of the eye, the anterior corner of the
eye, or slightly beyond instead of barely reaching the eye. Upper lip broadly
edged with white, posterior face of femur black reticulated with white, chest
and throat of young specimens washed with grey which may persist in the
adult ; — instead of upper lip narrowly edged with white, posterior face of
femur divided by a yellow line, ventral surface immaculate in both young
and adult.

Range. The wet, forested subtropical highlands ("montaiia")
from Tabaconas to Charape in northwestern Peru.

Type. M. C. Z. 5,344 from stream-bed at Tabaconas (near Huanea-
bamba) northwestern Peru; 1 September, 1916, G. K. Nol)le.

Description of Type. Snout moderately prominent, as long as the diameter
of the orbit; loreal region vertical; nostril nearer the tip of the snout than the
e3^e; interorbital space a little broader than upper eyelid; tympanum half
concealed by a glandular fold, about one third the diameter of the eye;
discs well developed, smaller than tympanum ; two small metatarsal tubercles,
inner oval, outer rounded; a small oblique ridge in the middle of the inner
side of the tarsus; tibiotarsal articulation reaching to the middle of the eye.
Skin smooth, glandular on the sides.

Uniform dark brown above; a black stripe extending entirely around the
body; the upper edge of this stripe, in the body region, bordered with yellow

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 397

and the lower edge reticulated with white. Upper lip and lower part of the
face, from a line connecting the nostril and tympanum, fleshy white; limbs
pale brown blotched with dark brown; liinder sides of thighs reticulated with
white; lower parts and ventral surface of body pinkish white, immaculate.

Dimensions.

Distance from snout to vent 33 mm.

Greatest width of head 11

Distance from axilla to tip of longest digit 19 "

Distance from groin to tip of longest toe 46 "

Notes on Paratypes. The twelve other specimens of the series
range in size from fifteen to thirty-two millimeters (snout to vent).
One specimen still possesses a large part of the tail. The small speci-
mens are very different in coloration from the adult. Instead of the
ventral surface being immaculate there is present a heavy wash of
grey extending completely over the throat, chest, and anterior part
of the abdomen. In most of the small specimens this grey wash is
finely spotted with white. The young of P. latinasus do not have
the grey wash. iVt least they are not so recorded. One specimen
(M. C. Z. 2,899) of P. latinasus from Chimbo, Ecuador is of nearly
the same size as two of our small specimens and yet there is no trace
of the gre}' wash. In two of the large parat^^es of P. sylvatica, both
males, this wash is present but only faintly indicated, while it is entirely
absent in all of the adult females.

There are only two adult males in the collection. These differ from
all the other specimens in being weakly tubercular above. x\pparently
in this species as in B^ifo mar inns — to a greater extent — rugosity is a
secondary sexual character. It has been shown that the males of
certain other species of Phyllobates transport the larvae on their
back. Two tadpoles of this species w^ere taken in a small pond.
These do not differ materially except in size from the tadpoles of the
tadpole-carrying-species, P. subpundatus {cf. Ruthven and Gaige,
Occas. papers, Univ. Mich., 1915, no. 10). Still in the two tadpoles
of P. sylvatica the second row of teeth is divided by a much shorter
interspace than in the tadpoles of P. subpunctatus. If the male of
P. sylvatica does carry its larvae, it is possible that the rugosity may
help the tadpoles to maintain their hold.

Besides the presence or absence of the ventral wash of grey, there
are other variations in the adults. The dorsal surface may be brown.

398 bulletin: museum of comparative zoology.

spotted with black. In that case the lateral yellowish stripe is very-
wide, and the dark cross-bars on the thighs are in sharp contrast to the
greyish ground-tones. Most of the adults have some dark spots on
the back, and most have the dark lateral band profusely reticulated
with white.

Phyllobates infraguttatus Boulenger.

Nearly a thousand specimens from several localities: — Palarabla
and Huancabamba (August), Perico and Bellavista (September), and
Querocotilla (October).

We have not been able to find any character with which to dis-
tinguish the specimens from Palambla, Huancabamba and Queroco-
tilla from a specimen (M. C. Z. 3,214) of P. infracjiittatus taken at
Rio Chanchan, Ecuador. The majority of our specimens from the
mountains of northern Peru are dark grey below spotted with white,
similar to the Ecuadorian specimen of P. infraguttatus before us, but a
few of those from the movmtains and all of those from the lowlands
(Perico and Bellavista) have the ventral surfaces white with a pair
of large dark spots just anterior to the pectoral girdle and sometimes
a delicate marbling of the same tone along the sides of the belly. The
pair of dark spots is scarcely visible in the dark bellied specimens from
the mountains and no such spots are mentioned in Boulenger's descrip-
tion of P. infraguttaius. Nevertheless they are apparently invariably
present in some degree and form the most important distinguishing
character of the species.

Peracca (Boll. Mus. Torino, 1904, 19, no. 465, p. 17) has shown that
Phyllodromus must be referred to Prostherapis, a genus indistinguish-
able from Phyllobates as defined by Boulenger (Proc. Zool. soc. London,
1888, p. 206). The notch on the posterior margin of the tongue is
extremely variable in all the species of Phyllobates and Prostherapis
which we have examined. In Phyllobates trinitatus the tongue is often
entire, while in our huge series of P. infraguttatus the tongue is some-
times emarginate and sometimes entire. In the four specimens of
P. latinasus before us the notch shows various degrees of de\elopment.
We have examined specimens of Prostherapis inguinalis and P.
boulengeri having an emarginate tongue. It is evident then that the
emarginate tongue does not distinguish Phyllobates from Prostherapis.

In internal structure Prostherapis agrees essentially with Phyllo-

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 399

bates. The type of the latter genus, P. hicolor} is known only from
the tj^e-specimen. This is at present in the Museum d'Histoire
Naturelle in Paris, and thanks to the kindness of Dr. Louis Roule
we are able to make some statement as to its internal structure. Its
pectoral girdle is similar to that of PhyUobates trinitatus, P. pratti,
P. infraguttatus, Prostherapis inguinalis, and P. boulengeri. The girdle-
form of P. trinitatus Garman (Fig. 1) may be taken as characteristic

Fig. 1. — PhyUobates trinitatus Garman. Shoulder-girdle, x 18.

of all these species. In the sketch of the girdle of P. hicolor sent us
by Dr. Roule, the anterior part of the omosternum is not represented.
It may have been lost during dissection. The fact that the form of
the coracoids and metasternum agrees in all the species examined is
excellent indication of the species close affinity. We can see no reason
for maintaining Prostherapis distinct from PhyUobates.

There remains another genus which should be referred to the
synonymy of PhyUobates. Colostethus, known only from the type-
specimen of C. latinasus, was described by Cope as lacking the omo-
sternum. The type-specimen cannot be found in either the U. S.
National Museum or the Academy of Natural Sciences of Philadelphia.
Specimens very similar to C. latinasus, but possessing an omosternum,
have been found in Ecuador and Colombia. Boulenger has recently
advised the senior author by letter that he considers his PhyUobates
pratti referable to C. latinasus. Boulenger evidently considers that
Cope overlooked the omosternum, a very probable oversight, in view

' The locality of this species has been discussed by the senior author (Mem. M. C. Z., 1919,
47, p. 113).

400 bulletin: museum of comparative zoology.

of the smallness of the type-specimen. Since the type of C. latinasus
is most probably lost, we consider it advisable to follow Boulenger's
suggestion and to place Colostethus in the synonymy of Phyllobates.

The number of species of Phyllobates is greatly increased by the
referring of Phyllcdromus, Prostherapis, and Colostethus to its syn-
onymy. It is highly probable that some of the species are not actually
distinct from others. The name P. pvklicllus is based upon a figure
only (Espada, Vertebrados del Viaje al Pacifico, Batracios, 1875, pi. 3,
fig. 3, 3a, 3b, and 3c) of a frog, without data. Unfortunately the
figures do not show the ventral surface. Seven species of Phyllobates
are described with a dorsal pattern very similar to that represented
as P. i^nlchdhis. These seven species are most readily distinguishable
by their ventral coloration. Any one of them may be referable to
P. j^ulchellvs but without a description of their ventral surfaces it is
impossible to say which of them should bear the name. Our speci-
mens of P. infra gidfatns are very similar to Espada's figure. More-
over, we have recently examJned some specimens (U. S. N. M. 61,763,
61,764) of the same species from Huigra, Ecuador. These offer further
evidence of the wide distribution of the species in Ecuador and north-
ern Peru. It is probable that Espada's t^pe of P. jndchcllus came
from Ecuador. But until this type-specimen can be examined and
sufficiently characterized we have no right to refer Boulenger's P.
infragidfatus to that species. The name P. jmlchcUus must be dropped
until the type-specimen can be examined.

We have remarked about the similarity of color-pattern in several
species of Phyllobates. The dorsal color-pattern shows little variation
in our enormous series of P. mfragiittahis. The ground-tone may be
light or dark grey. The pale specimens, m.ostly from the lowlands,
have the dorsal spotting very distinct and in sharp contrast to the
light grey ground-tone. Dark specimens are uniform above, the spots
being concealed by the ground-tone. The characteristic pair of spots
on the chest are indistinct in the darkest and practically wanting in
the very lightest specimens of the series.

At Querocotilla one specimen was found which appeared nearly a
uniform pea-green. The specimen was put in a jar of fixing fluid
(.5% formol) and in the course of half an hour the green tones were
replaced by greys and the ordinary pattern appeared making the speci-
men indistinguishable from the others.

The species of Phyllobates are all small, and possess few features
by which they may be distinguished from one another. Still it has
been considered advisable to attempt a key, unfortunately based

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 401

largely upon the published descriptions. In the preparation of this
key certain species have been found to be certainly synonymous with
others. Prostherajns equatorialis Barbour is referable to Eleuthero-
dadylus unistrigatus (Giinther), Prostherapis herminae Boettger to
PhyUobafcs trinitatus Garman, Prostherapis variahilis Werner to
PhyUobatcs suhpunctahis Cope, and Phyllobates {Hypodidyon)
palmatus Werner to Syrrhophus palmatus Werner.

Key.

A. Flash colors of red or yellow on the axilla and inguinal or femoral
regions; dorsal surface of head and body distinctly glandular.
B. First finger longer than second.

C. Flash colors yellow femoralis (Boulenger).

CC. Flash colors pink ing^iinalis (Cope).

BE. First finger not extending beyond second.

C. Ground-tone of back lemon-yellow tricolor (Boulenger).

CC. Ground-tone of back brownish grey festae (Peracca).

AA. No flash colors; dorsal surface of head and body not distinctly glandular
although sometimes tubercular.
B. Lower surfaces of adult white, immaculate.

C. Tibiotarsal articulation not reaching beyond tympanum.

vertebralis (Boulenger) .
CC. Tibiotarsal articulation reaching to eye.

D. Skin with numerous low, glandular warts, tympanum concealed.

briinneus (Cope).

DD. Skin smooth or with scattered tubercles, tympanum indistinct.

E. T\Tnpanum one half diameter of eye; tibiotarsal articulation

reaching barely to eye latinasus (Cope) .

EE. Tympanum one third diameter of eye; tibiotarsal articulation
reaching to middle of eye or beyond.

sylvatica Barbour & Noble.
BB. Ventral surface of adult brown or spotted with dark tones.

C. A dark bar, two dark spots or a dark wash spotted with white on
the chest.
D. A dark bar across the chest, anterior part of belly generally of

same dark tone trinitatus Garman.

DD. No dark bar across chest.

E. Throat and breast mottled with greyish brown.

kingsburyi Boulenger.
EE. Ventral surface dark, spotted with white, sometimes only two
greyish spots on chest.
F. Black above, spotted with white on sides.

alboguttatus Boulenger.

402

bulletin: museum of comparative zoology.

FF. Greyish above, no spots on sides, generally a longitudinal

stripe on each side infraguttatus Boulenger.

CC. No dark bar or spots on chest.

D. Ground-tone of ventral surface uniform blackish.

rnelanorhinus Berthold.
DD. Ground-tone of ventral surface greyish or whitish.
E. Tympanum concealed, throat greyish, belly white.

trilineatus Boulenger.
EE. Tympanum not concealed.

F. Tympanum two thirds diameter of eye.
G. Two white lines on each side of head.

bolivianus (Boulenger).

GG. No white lines on head whymperi (Boulenger).

FF. Tympanum about half the diameter of eye, or less, sometimes

indistinct.

G. First finger longer than second; tympanum hidden or

about one third diameter of eye. . . . boulengeri (Barbour).

GG. First two fingers of equal length; tympanum about half

as broad as eye siibpunctatus (Cope).

One species, generally referred to Phyllobates, differs radically
from all the species of that genus and must be considered generically
distinct. We propose for it the name :

Sminthillus, gen. nov.

Type. Sminthillus limbatus (Cope).

Diagnosis. Habit of Phyllobates but no pair of dermal scales on
the upper surface of the digital discs; coracoids narrow (Fig. 2);

Fig. 2. — Sminthillus limbatus (Cope). Shoulder-girdle, x 27.

precoracoid cartilages very broad, merging gradually into the broad
epicoracoid cartilage; omosternum cartilaginous, without a bony

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 403

sheath in the adult; sacral diapophysis slightly dilated; no vomerine
teeth; tongue elliptic, narrow and free behind; pupil horizontal;
tympanum distinct; toes free; terminal phalanges T-shaped.

Leptodactylidae.

Eleutherodactylus lymani/ sp. nov.

Diagnosis. Similar to E. conspicillatus (Giinther) in most of its features,
but differing from that species in the much shorter toes, especially the fourth
toe; in the longer first finger and in the somewhat different color-pattern.
Young specimens are similar in habit to that species but adult specimens are
much stouter and with a broader, less acuminate snout.

Range. Hills and valleys of the central Andes of northwestern Peru,
from Palambla (near Huancabamba) to Bellavista.

Tyjie. M. C. Z. 5,422 from Perico, valley of the Chinchipe, north-
western Peru; 10 September, 1916, G. K. Noble.

Description of Type. Size large, head broader than body, exactly as long
as broad; snout subacuminate with distinct canthus rostralis and concave
loreal region; orbital diameter equals the distance between the eye and nostril,
twice as great as the distance from nostril to end of snout; interorbital space
a trifle broader than the upper eyelid. Tongue oval, slightly nicked behind.
Vomerine teeth in two oblique, approximated fasciculi behind the choanae.
Tympanum distinct, half the diameter of the eye. Fingers moderate, the first
extending beyond the second; toes short with a rudiment of a web; discs
small, not much wider than the middle part of the toes; subarticular tubercles
well developed; a round outer and an elongate inner metatarsal tubercle, both
very distinct. Tibiotarsal articulation reaches just to the tip of the snout.
Sk'in finely granular on the back, nearly smooth on the head, and coarsely
granular on the lower surfaces of the thighs; no glandular dorsolateral fold.

Ground-color above, ashy grey fading to yellowish grey on the sides. Two
che\Ton-shaped bands of dark brown on the back, one over the scapulae and
one just before the ilia; two or three spots of the same color posterior to the
ilia. A narrow stripe of dark brown along the canthus rostralis and over the
tympanum; a faint interorbital bar. Lips and appendages cross-barred with
dark brown; four bars across the legs; posterior surfaces of thigh reticulated
with dark brown and white. Ventral surface of head and body white, un-
spotted; of feet a dark brown.

1 Named in honor of Prof. Theodore Lyman whose generosity enabled the Museum of Com-
parative Zoology to send a zoologist with the Harvard Peruvian expedition of 1916.

404 bulletin: museum of comparative zoology.

Dimensions.

Distance from snout to vent 52 mm..

Greatest width of head 21 "

Distance from axilla to tip of longest finger 29 "

Distance from groin to tip of largest toe 89 "

Notes on Paratypcs. The twenty-nine specimens in the series show
considerable diversity in color. The ground-tone varies from a yellow
or a dull pink to a very dark brown. The specimens vary in length
from 23 to 58 millimeters (snout to vent). Most of the small speci-
mens have a very distinct pattern with a dark interorbital bar and two
spots anterior to it. Two small* specimens have a light vertebral line.
The pattern does not appear in a few of the specimens. Very dark
specimens have the throat stippled with brown.

Eleutherodactylus cajamarcensis, sp. nov.

Diagnosis. Related to the Borborocoetes group of Eleutherodactylus, e. g.
E. whijmperi, E. unistrigatus etc.; probably most closely related to E. riveti
(Despax) ; distinguished from the latter species by the following characters : —
tympanum distinct slightly less than half the diameter of the eye; first toe
shorter than the second; skin smooth on the snout, slightly granular on the
eyelids and back, the granules tending to form a series of longitudinal rows;
coloration nearly uniform yellowish grey; a few dark lines forming a weak
pattern; ventral surface immaculate.

Range. Only known from the type-specimen.

Tyjye. Sexually mature male, M. C. Z. 5,407 from the Pre-Incan
ruins near Huambos, Cajamarca, northwestern Peru; 10 October,
1916, G. K. Noble.

Description of Type. Size small, body depressed; head broad, Hyla-like;
head about as broad as the body; broader than long; snout blunt with very
distinct cmithus rosiralis; orbital diameter much greater than the distance
between eye and the nostril; the latter situated very near the end of the snout;
interorbital space a little broader than upper eyelid. Vomerine teeth barely
distinct, in two obliquely directed groups, extending backward from the inner,
posterior margins of the choanae. Tympanum distinct, slightly less than half
the diameter of the eye. Fingers stout, the second extending beyond the first;
toes short, a vestige of a web between the outer three; discs distinct, those
of the toes larger than those of the fingers; subarticular tubercles well de-

BAKBOUR AND NOBLE: AMPHIBIANS FROM PERU. 405

veloped, a round outer and a very elongate inner metatarsal tubercle, both
very distinct. Tibiotarsal articulation reaching only to the tympanum.
Skin smooth on the snout, slightly granular on the eyelids and back, the
granules on the back tending to form a series of indistinct longitudinal rows;
sides of the body warty; ventral surface strongly granular.

Coloration in alcohol nearly uniform yellowish grey; a dark canthal stripe
fading out behind the tympanum; a number of indistinct brownish bands
extending along the back; three oblique bands across the legs, these tending
to form continuous lines when the leg is half extended; ventral surface uni-
form yellowish grey, much yellower than the dorsal surface. In life the
ground-tone was yellowish pink and the dark pattern was fairly distinct.

Dimensions.

Distance from snout to vent 19 mm.

Greatest width of head 7.5

Distance from axilla to tip of longest digit 12

Distance from groin to tip of longest toe 26 . 5

Leptodactylus curtus, sp. nov.

Diagnosis. A short-legged species having no fringes on the toes, apparently
related to L. bufonius Boulenger; head short, the profile chisel-shaped; tym-
panum half the diameter of the eye; no distinct dorsolateral fold; back and
sides with a few low warts.

Range. Valleys of the Chinchipe and Maraiion Rivers between
Perico and Bellavista, northwestern Peru.

Type. M. C. Z. 5,281 from Bellavista, Cajamarca, Peru; 28 Sep-
tember, 1916, G. K. Noble.

Description of Type. Size moderate; head about as wide as the body, just
as long as broad; snout very accuminate without canihus rostralis, but with a
slight depression in the loreal region; profile of snout a very acute angle, the
anterior corner of the eye, the nostril and the tip of the snout being in the same
plane; orbital diameter slightly greater than the distance between eye and
nostril, slightly less than the distance between nostril and end of the snout;
interorbital space about one half as broad as the upper eyelid. Tongue oval,
slightly nicked behind. Vomerine teeth in two well-arched series behind the
choanae. Tympanum one half the diameter of the orbit. First finger much
longer than the second; toes short, not fringed; subarticular tubercles well
•developed; the inner metatarsal tubercle very large, the outer barely visible;

406 bulletin: museum of comparative zoology.

a distinct tarsal fold. Tibiotarsal articulation reaches to the tympanum.
Skin glandular but not tubercular; a few flat warts on the back and sides;
these tend to form a weak dorsolateral fold; a large ovoid gland situated at
each corner of the mouth, its posterior end directed downward; a large inguinal
gland on each side of the body; a small but distinct glandular wart on the
posterior face of each femur.

Ground-tone of dorsal surface olive-grey; a number of dark brown spots
forming a pattern ; an hour-glass-shaped figure between the eyes and pectoral
region; the anterior end of the figure much wider than the posterior, and out-
lined with pale grey; the posterior end of the figure continuous with two
rows of dark spots which extend the length of the back; a series of dark spots
along the side of the head and body; these tend to form a line along the
indistinct dorsolateral fold; two or three spots on the lips; the legs irregularly
cross-barred; the posterior surfaces of the thighs reticulated with black and
white; ventral surface whitish, immaculate.

Dimensio7is.

Distance from snout to vent 49 mm.

Greatest width of the head 19 "

Distance from axilla to tip of longest digit 26 "

Distance from groin to tip of longest toe 66 "

Notes on Paratypcs. The series of twenty-tw'o specimens shows a
great uniformity in the proportions of the body. The color-pattern is
subject to some variation. This consists chiefly in a multiplication of
the dark spots, and in a fading or intensifying of the ground-tone.
Very dark specimens have the periphery of the ventral surface stippled
with dark brown.

Remarks. A study of a series oi Lepiodactylus albilabris from St.
Croix has led to the conclusion that the chisel-shaped head form of
L. curtus may be only a somatic variation. All of the specimens in
our series of the latter species (these measuring in length, snout to
vent, from forty to fifty-eight millimeters) have exactly the same
head-form regardless of sex. It has been suggested (Barbour, Proc.
Biol. soc. Wash., 1917, 30, p. 103) that this head-form might be a nup-
tial modification; it does not seem that it is a secondary sexual char-
acter, at least not in the case of L. mrius. In our specimens even the
youngest are hardly sexually mature. Since these were all taken
around sandy sloughs, it is possible that the head-form may be a modi-
fication for burrowing in the sand. Direct field observation is lacking
on this point.

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. . 407

Telmatobius Wiegmann.

The status of the genus Tehnatobius has not been understood. Its
true relations cannot be determined until the internal structure of
the type, T. peruviamis, has been described and the statements of Cope
(Bull. 34, U. S. N. M., 1889, p. 312) confirmed. For the present we
refer his genus Cophaeus to the synonymy of Telmatobius.

Telmatobius has been confused also with Cycloramphus. Speci-
mens of that genus are not available for study but judging from the
published descriptions the genus is a well-defined one. It is dis-
tinguished from Telmatobius by the presence of stout vomerine teeth
arranged in two long rows behind, not between, the choanae. Inguinal
glands are present in three of the four described species but are not
mentioned in C. hrasilicnsis (Steindachner). In Telmatobius not a
single species is so provided. The males of the latter genus, unlike
the former, are provided during the breeding season with dense asperi-
ties on the chest, forearm, and thumb. Boulenger (Cat. Batr. S^l.
Brit, mus., 1882, p. 184) distinguishes Cycloramphus from Telma-
tobius by its separated outer metatarsals. This character is not
mentioned in several of the descriptions, and until specimens can be ex-
amined it seems advisable to use the teeth characters as distinguishing
Cycloramphus from Telmatobius. In doing this we find that we have
two natural assemblages, Telmatobius confined to the Andes and
Bolivian Chaco and Cycloramphus to the highlands of Brazil. After
referring Tiimatohius hrasilicnsis '^temds.chnev 2iXi<\ T. (/ii^eni Andersson
to Cycloramphus, and upon placing T. asper Boulenger in the syn-
onymy of C. asper Werner we have four species of Cycloramphus which
may be separated by the following

Key.

A. Toes less than half webbed duseni (Andersson) .

AA. Toes more than half webbed.

B. Skin smooth fuliginosus Tschudi.

BB. Skin warty or provided with horny tubercles.

C. Skin of the body loose and wrinkled; toes completely webbed.

brasiliensis (Steindachner).
CC. Sldn of body not loose dorsally; toes not fully webbed.

asper Werner.

408 bulletin: museum of compakative zoology.

The species of Telmatobius are so little known that it is advisable
to redescribe the type-specimens of several of the species and to
append a key of all those considered distinct. In reviewing the species
the fact has presented itself very forcibly that Telmatobius within
itself represents various stages in the reduction of both maxillary and
vomerine teeth. This reduction of teeth is associated with aquatic
life. In the Lake Titicaca region T. acmaricus is found along the
■edges of the small streams and ponds, while T. culeus occurs only in
the deep waters of Lake Titicaca where according to Garman (Bull.
M. C. Z., 1875, 3, p. 277) it is able to remain for hours without coming
up to breathe. Garman (Loc. cit.) says "As might be expected from
the exclusively ac^uatic habits of culeus, its skeleton is weaker and
less perfectly ossified than that of marmoratus [= our aemaricus].
In the latter the skull and its processes are strong and the foramina
:and fontanel very small."

Garman pointed out that the vomerine teeth were very reduced,
sometimes absent on one side or the other. We have found that the
maxillary teeth of T. culeus are also much reduced in size.

As association exactly similar to that of T. culeus and T. aeviaricu^
is found in the Lake Junin region where T. jclsJcii is the semiaquatic
and Batrachophyrnus microphthalmu^ the lake-form. The latter species
although currently placed in a different genus and family from T.
culeus agrees entirely with it in most of its internal and external char-
acters. Peters (Monatsber. Akad. wiss. Berlin, 1873, p. 413) and
Werner (Abh. Zool.-anthro. mus. Dresden, 1901, 9, no. 2, p. 13, fig.)
have shown that Batrachophrynus is a Telmatobius in every particular
except that it lacks the maxillary and vomerine teeth. A comparison
of Batrachophrynus wuth T. culeus suggests that this difference is not
fundamental or of any real significance.

In the appended descriptions we have included Philippi's T. mon-
tanus and T. laevis. We are strongly inclined to follow Boulenger's
suggestion (Zool. record. Rept., 1902, p. 14) and disregard these
names as well as those of the other utterly uni'ecognizable species
which he has proposed (Supplementa a los Batraquios Chilenos
descritos en la Historia fisica y politica de Chile de don Claudio Gay.
Santiago, 1902). Nevertheless it seems highly probable that Philippi
had some species of Telmatobius before him when he wrote his paper,
so for the present it may be better to consider his proposed species
vahd.

Andersson's record (Ark. zool., 1906, 3, no. 12, p. 4) of T. jelskii
from the Andes of western Argentina and Werner's report (Zool.

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 409

jahrb. Suppl., 1897, 4, p. 263) of T. aemaricus and T. marmoraius
in Chile certainly require confirmation. The status of the genus in
Chile is not at all clear.

Key.

A. Skin warty, at least provided with strong tubercles.
B. Vomerine teeth in two small groups.

C. Tibiotarsal articulation reaching the tip of the snout; color above

brownish grey spotted with dark brown. . . .verrucosus Werner.

CC . Tibiotarsal articulation reaching the angle of the mouth ; color above

uniform blue-gray hauthali Koslowsky .

BB. Vomerine teeth absent or barely visible.

C. Entire upperside of legs provided with tubercles, a distinct color-
pattern peruvianus Wiegmann.

CC. Tubercles on the legs restricted to the dorsal side of the tibia, no

distinct color-pattern pustulosus (Cope).

AA. Skin smooth or granular.

B. A strong supratympanic fold.

C. Vomerine teeth prominent, in two large groups.

D. Tympanum hidden niger Barbour & Noble.

DD. Tympanurn present ignavus Barbour & Noble.

CC. Vomerine teeth present or absent, generally in two very small groups.
D. Skin very loose, large femoral flaps; tibiotarsal articulation
reaches only to the corner of the mouth. . . .cu'eus (Garman).
DD. Skin loose only on the sides.

E. Ventral coloration uniform pale below aemaricus (Cope).

EE. Ventral coloration sooty grey pale on throat only.

jelskii (Peters).
BB. No supratympanic fold.
C. Vomerine teeth absent.

D. Toes fully webbed monlanus Philippi.

DD . Toes one quarter webbed laevis Philippi.

CC. Vomerine teeth present.

D. Tibiotarsal articulation not reaching the eye; color above brown

mottled with black ". .marmoraius (DumSril & Bibron).

DD. Tibiotarsal articulation extending to the anterior edge of the eye;
color uniform brown above angustipes (Cope).

Telmatobius VERRUCOSUS Wemer.

Telmatohius verrucosus Werner, Zool. anz., 1899, 22, p. 482.

Diagnosis (extracted from original description). Vomerine teeth in two
round groups between the choanae. Tympanum hidden. Toes webbed to the

410 bulletin: museum of compaeative zoology.

basis of the last phalanges, only in the fourth toe webbed to the base of the
next to the last phalange; the free phalanges of the toes with a distinct seam
in the skin, especially distinct on the outer edge of the fifth toe. Tibiotarsal
articulation reaches the tip of the sncut. Upper surface rugose, upper lip,
appendages, and urder surface smooth. A strong fold from the posterior
edge of the eye to the corner of the mouth. Color above brownish grey,
spotted with dark brown, the tubercles all dark brown. Ventral surface
uniform light greyish brown.

Habitat. "Chaco, Bolivia." The species is known only from the
original description.

Telmatobius hauthali Koslowsky.

Tdmaiolius hcuihali Kcslcwsky, Revista Mus. La Plata, 1895, 6, p. 359, pi. 1.
{7)TeIma{obius jelsMi Andersson, Arkiv. zool., 1906, 3, no. 12, p. 4.

Diagnosis (extracted frcm original description). Vomerine teeth in two
small groups between the chcanae. Toes more than half webbed, a free border
extending along the interior edge of the tarsus, and on the outer side of the
interior toe. JThe leg extended forward reaching with the tibiotarsal articula-
tion the angle of the mouth. Skin smooth, covered with numerous horny
tubercles on the chest, back, abdomen, and appendages. Lead color or blue-
grey above, darker on the head region; ventral suiface dirty yellowish white;
some specimens mottled with dark tone on the posterior part of the ventral
surface, all specimens uniform above.

Habitat. The type-locality: Andes of Catamarca, Argentina,
Aguas Calientes, a streamlet 4,060 meters above sea-level.

Remarks. Oddly enough this species has been found only in the
warm waters of a hot spring. In the type description Koslowsky
(1895, p. 360) says: — ^"Senor Rodolfo Hauthal los tomo en el arroyo,
cuyas aguas siempre conservan una temperatura de veinte grados,
Celsius." How different from the frigid waters of Lake Titicaca,
the home of the closely related T. culeus !

Telmatobius peruvianus Wiegmann.

Telmatobius -peruvianus Wiegm., Nova acta, 1835, p. 262, pi. 22, fig. 2.
Telmatobius peruvianus Boulenger, Cat. Batr. Sal. Brit, mus., 1882, p. 191.

Diagnosis (extracted from the original description) . Presence of vomerine
teeth questionable. Toes with short webs. Skin finely granular, covered

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 411

upon the head, the whole back and the upper surface of the legs with many
small flattened tubercles, each of which is provided with a hard, horny point,
dark in color. Skin of under surface smooth except for the characteristic
horny tubercles in the pectoral region characteristic of the male Telmatobius,
in the breeding season. Color above brown, a darker band extending from
the snout to the foreback forming with two cross-bands drawn from the ear
to the shoulder region a feeble double cross; under surface of the body and
legs a brown-yellow.

Habitat. Cordillera de Guatilla, near the town of Palca, two days'
journey east of Tacna, Chile (Meyen).

Remarks. It is highly probable that T. peruvianus is a land-frog
with habits similar to those of T. aemaricus, for Meyen says in his
account of finding the type: — •

"Unser Nachtlager schlugen wir einer natiirlichen Hohle des dicht
daneben anstehenden Gesteines auf;...gegen Abend Hess sich das
Quaken eines Frosches horen." (Reise um die erde, 1834, p. 450).

Telmatobius pustulosus (Cope).

Cyclorhamphus pustulosus Cope, Proc. Amer. philos. soc, 1877, 17, p. 39.
Telmatobius pustulosus Boulenger, Cat. Batr. Sal. Brit, mus., 1882, p. 192.

Diagnosis. A rather large species, the skin not especially loose but with
tubercles present on sides, belly, lower side of forearm, vent region, and lower
back and upper surface of tibia and sole of foot; vomerine teeth barely visible;
no distinct color-pattern.

Habitat. This was another of Professor Orton's discoveries during
his exploration of the Peruvian Andes. He secured the type and only
specimen known or recorded, at Tinta, a small town at an elevation of
11,400 feet in the Department of^Cuzco, Southern Peru.

Type. Acad. nat. sci. Phil., 11,401; Tinta, Peru; altitude 11,400 ft.

Description of Type. Size moderate ; head broader than long,»much broader
than body, its length contained in the total length of body 3.2 times; snout
rounded with no distinct canthus rostralis; nostrils although at the end of the
superior plane of the muzzle, equidistant between the orbit and the labial
border. Vomerine teeth barely visible, in two very small groups between the
choanae which are much enlarged; tongue small, nearly round. Interorbital
space 1.2 as broad as the length of the eye; the length of the snout 1.4 as long
as the length of the eye; tympanum concealed by the skin, but on one side

412

bulletin: museum of comparative zoology.

indicated by a vertical oval, the greatest diameter of which is one half the
length of the eye. Digits free, slender, without discs; the first and second
fingers equal in length and shorter than the fourth; the elbow extended for-
ward reaches a little beyond the orbit. Toes slender, without discs, less than

Fig. 3. — Telmatobius pustulosus (Cope). 1. Foot. 2. Hand. 3. Open mouth.

half webbed; subarticular tubercles only moderately developed; a distinct
inner metatarsal tubercle but no outer; heels barely in contact when the hind
limbs are folded at right angles to the axis of the body; the tibiotarsal articula-
tion reaches the anterior edge of the orbit when the hind limbs are carried
forward along the body. Skin smooth except for the low tubercles on the
sides of the body, underside of forearms, posterior dorsal surface of body,
region around the vent, dorsal side of the tibia, and ventral side of the foot;
the tubercles on the sides of the body much larger than the others, and more
whitish in color. No glandular folds on the dorsal surface, a few longitudinal
creases on the ventral surface.

Color above brown with a few indistinct darker markings ; light brown below,
gular region and lateral tubercles milky white.

Dimensions.

Tip of snout to vent 58 . 5 mm.

Tip of snout to posterior end of mandibular bone 21 "

Greatest width of head 22

Fore leg to tip of longest finger 36

Hind leg to tip of longest toe 85 . 5

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 413

Telmatobius NIGER, sp. nov.

Diagnosis. A mediiim-sized frog, with finely granular skin; a strongly
developed supratympanic fold, the vomerine teeth well developed in two large
prominent groups and with the tympanimi hidden.

Habitat. Many years ago a series of seven or eight of these frogs
were given the senior author by Mr. R. L. Ditniars, to whom they were
presented by a mining or railway engineer who had been to Ecuador
probably working on the Quito-Guayaquil Railroad at that time under
construction. They bore a label "Palmira Desert, Ecuador, 10,500
feet altitude." It has been impossible to identify this locality with
absolute certainty, but the Palmira is very probably the one referred
to as the hacienda de Palmira in the Andes of southern Ecuador by
Theodoro Wolf in his Geografia y Geologia del Ecuador (Leipzig,
1892, p. 35). The village of the hacienda is a few miles south of
Vilcabamba in the Province of Loja and is said to have an altitude of
1,748 meters. This is but half the altitude given on our label. There
may be another Palmira or the "desert" may in reality be a Paramo,
of the same name as the hacienda, not far away in the highlands
south of Loja.

Type. M. C. Z. 3,037 from Palmira Desert, Ecuador. Coll. Thomas
Barbour, 1909.

Description of Type. Size moderate; head broader than long, about equal
to the width of the body, its length contained in the total body length 3.3 times;
snout very short and high without canthus rostralis, nostril nearer the orbit
than the labial border. Vomerine teeth prominent in two well-defined groups
between the choanae which are of moderate size; tongue longer than broad.
Interorbital space 1.3 as broad as the length of the eye; the length of the snout
1.4 times that of the eye; tympanurp hidden, the region partly covered by the
supratympanic fold. Digits free, stout, slightly dilated at the tips, the first
finger longer than the second but shorter than the fourth; the elbow extended
forward reaches nearly to the eye. Toes fully webbed, the webs notched
making the toes appear only slightly more than half webbed; a distinct tarsal
fold; subarticular tubercles well developed; the inner metatarsal tubercle
much larger and more prominent than outer; heels not in <'ontact when the
hind limbs are folded at right angles to the axis of the body; the tibiotarsal
articulation reaches to the middle of the eye when the hind limb is carried
forward along the body. Skin very glandular above, the glands being so
small that the skin appears granular and not warty; a loose, baggy, latera
fold on each side, the fold beginning at the posterior angle of the eye and con

414

bulletin: museum of comparative zoology.

tinuing to the groin; in the suprat5rmpanic region this fold somewhat swollen
resembling on one side a parotid gland; skin on the sides of the body loose and
folded; ventral disc marked off by a transverse fold between the fore limbs

Fig. 4. — Telmatobius niger Barbour & Noble. 1. Foot. 2. Hand. 3. Open mouth.

and by two longitudinal creases, one on either side of the belly; skin on the
thighs loose, but no posterior flap.

Color above very dark chestnut-brown; ventral surface yellow, heavily
mottled and spotted with dark brown; the spots most abundant on the throat,
east so upon the thighs; tips of fingers and toes yellow.

Dimensions.

Tip of snout to vent 61 mm.

Tip of snout to angle of jaw 21 "

Greatest width of head 22 "

Fore leg to tip of longest finger 37 "

Hind leg to tip of longest toe 91 "

Remarks. The only specimens known are the typical series some of
which have been distributed under the name of Telmatobius jelskii
to the U. S. N. M. and possibly elsewhere. Until topotj'pes of jelskii
were secured recently, we had these two species confused.

Telmatobius ignavus, sp. nov.

Diagnosis. A medium-sized smooth-skinned frog with strongly developed
supratympanic fold, the vomerine teeth well developed in two large prominent
groups and with the tympanum exposed.

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU.

415

Habitat. Central Cordillera of Piura, northern Peru. It may
perhaps also be expected to occur in the higher portions of Cajamarca
and Lambayeque.

Type. M. C. Z. 4,093 within the town limits of Huancabamba,
Piura, Peru; 5 August, 1916, G. K. Noble.

Description of Type. Size moderate; head broader than long, much nar-
rower than the body, its length contained in the total length of the body just
three times; snout rounded and flat without canthus rostralis, nostril slightly
nearer the orbit than the labial border. Vomerine teeth in two large groups
between the choanae which are very smaU; tongue moderate in size, longer
than broad. Interorbital space 1.4 as broad as the length of the eye; the
length of the snout 1.5 times that of the eye; tympanum one fourth the diame-
ter of the eye, partly covered by the supratympanic fold. Digits free, stout,
slightly dilated at the tips, the first finger longer than the second but shorter
than the fourth; the elbow extended forward reaches slightly beyond the

Fig. 5. — Telmatobius ignavus Barbour & Noble. 1. Foot. 2. Hand. 3. Open mouth.

tympanum. Toes more than half webbed, but the webs notched, making the
toes appear only one third webbed; a narrow border of free skin On the inner
side of the tarsus; subarticular tubercles distinct; the inner and outer meta-
tarsal tubercles well developed; heels just in contact when the hind limbs are
folded at right angles to the axis of the body; the tibiotarsal articulation
reaches nearly to the posterior edge of the orbit when the hind limbs are
carried forward along the body. Skin smooth; upper surface of body, and
appendages somewhat glandular, the glands appearing as irregular patches of
pores; a well-developed lateral fold, beginning at the posterior angle of the
orbit and extending to the groin; skin on the sides of body very loose and baggy;
ventral disc marked off by a transverse fold between the fore limbs and two

416 bulletin: museum of comparative zoology.

longitudinal ones on either side of the belly; skin on the thighs somewhat
loose, but without posterior flaps.

Color above dark reddish brown faintly marked by large spots of a darker
color; ventral surface uniform cream-white mottled on the thighs with brown
and pure white.

Dimensions.

Tip of snout to vent 54 . 5 mm.

Tip of snout to posterior end of mandibular bone 20 "

Greatest width of head 22 "

Fore leg to tip of longest finger 32 "

Hind leg to tip of longest toe 72

Remarks. The type and only specimen secured was taken in a small
cistern at the western end of the town of Huancabamba. The junior
author was searching for Gastrotheca larvae in this cistern, not three
feet in diameter, when he suddenly espied this good-sized frog. None
of the boys present seemed to know it, for to his urgent " Covio se llama
esta Rana," they only shook their index-fingers in that peculiar wagging
manner by which the Peruvian " cholo " signifies that he does not know.
Diligent collecting in the streams about Huancabamba for a month
did not reveal another specimen.

Telmatobius culeus (Garman).

Cyclorhamphus culeus Garman, Bull. M. C. Z., 1875, 3, p. 276, pi.
Telmatobius jelskii (?) Boulenger, Cat. Batr. Sal. Brit, mus., 1882, p. 191 (pars).

Diagnosis. A very large species, with a smooth loose, baggy sldn; vomerine
teeth so reduced as to be almost invisible; strong supra tympanic, lateral and
femoral folds or lappets; the tibiotarsal articulation reaching only the angle of
the jaws.

Habitat. Confined to Lake Titicaca where it leads an absolutely
aquatic existence.

Type. M. C. Z., 1,077 from bottom of Lake Titicaca, Peru; taken
dredging by S. W. Garman, in eleven fathoms off Achacache, Bolivia.

Description of Type. Size very large; length of the head contained in the
breadth 1.5 times, in the total length 3.8 times; snout round, very flat without
canthus rostralis; nostril minute, slightly nearer the orbit than the labial
border. Vomerine teeth reduced to a few very small spines which scarcely
break through the buccal epithelium at two points between the choanae; the

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU.

417

choanae enlarged and at a slight angle to each other; tongue small, longer than
broad. Interorbital space nearly two times as broad as the length of the eye;
tympanum hidden, the region covered by two loose flaps of skin. Digits
slender but edged on either side by a seam of sldn which does not develop into
a web, tips not dilated; the first finger a trifle longer than second, equal in
length to the fourth; the elbow extended forward reaches only about three

Fig. 6. — Telmatobius culeus (Garman). 1. Foot. 2. Hand. 3. Open mouth.

fourths of the distance to the orbit. Toes fully webbed but deeply notched,
making them appear slightly more than half webbed; a free border of skin on
the inner side of the tarsus and a narrower one on the outer edge of the fom-th
toe; subarticular tubercles barely visible, the metatarsal tubercles the most
distinct; heels almost in contact when the hind limbs are folded at right angles
to the axis of the body; the tibiotarsal articulation reaches slightly beyond the
angle of the mouth, when the hind limb is carried forward along the body.
Skin smooth, the entire upper surface glandular, the glands appearing as
numerous small depressions; skin very loose and baggy, falling into a number
of large folds and flaps; a supraocular fold continued posteriorly to form a
supratympanic and lateral fold, several transverse folds on the sides; a very
wide, loose, liquid-filled bag or flap extending along the posterior side of each
thigh; many folds on the appendages due to the loose character of the skin.

Color slate-grey above thickly spotted with white; the spots more abun-
dant ventrally giving the under surface a lighter appearance.

418 bulletin: museum of comparative zoology.

Dimensions.

Tip of snout to vent 97 mm.

Tip of snout to posterior end of mandibular bone 36 "

Greatest width of head 43 "

Fore leg to tip of longest finger 52 . 5 "

Hind leg to tip of longest toe 135 "

Remarks. Mention has already been made of the probable evolu-
tion of T. culcus from T. aemaricus. Garman is the only herpetolo-
gist who has ever published upon the peculiar subaqueous existence
of T. cideus. In the original description he says : —

"These animals are very abundant in the extensive beds of weeds
which occur on the bottom of Lake Titicaca. They feed on the mol-
luscs, Crustacea, worms, etc., and are fed upon by birds and fishes.
Marmora tus [= aemaricus] was found in little creeks and marshy places,
in situations indicating habits similar to the common Ranpe; during
the two months of the observations culeus was only to be fovmd in the
lake, crawling lazily about among the weeds or half hidden by them,
watching for prey. The latter was the only one found in the vicinity;
the former was secured on the summit and the western slope. These
animals are able to remain vmder for great lengths of time without com-
ing up for air; hours of watching in clear water where many could be
seen, failed to detect any approaching the surface. It is possible that
they are more lively at night, when their enemies are less active.
Numbers were brought up in the trawl at more than four miles from
the shore. None were found on the land. The natives were positive
they never left the water. All stages of the animal are represented
by the specimens in the collection."

Mr. Garman in conversation, adds the following facts : — the species
is far more abundant in the southern than in the northern end of the
Lake. The large specimens obtained were mostly caught near Guaqui
or Copacabana. None were observed or dredged in the great northern
bay north of the Capachica Peninsula. Neither Garman nor the
senior author, who visited Lake Titicaca in 1909 found that the
Indians used culeus for food. This is a common custom at Lake
Junin with Batrachophrynus micro phthalmus, whose habits are very
similar to those of the Titicaca frog.

The following material in addition to the type remains in the Mu-
seum of Comparative Zoology, of Garman's collecting, after a consider-
able number of specimens have been distributed to other institutions.

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU.

419

2 veiy large adults from Copacabana, Bolivia.

1 half-grown specimen from Lake Titicaca, (no further data).

5 half-grown specimens from Carapata Bay, Bolivia.

6 half-grown specimens from Desaguadero Bay, Bolivia.
1 half -grown specimen from Puno, Peru.

8 larvae from Puno, Peru.

Telmatobius aemaricus (Cope).

Cyclorhamphus aemaricus Cope, Proc. Acad. nat. sci. Phil., 1874, p. 125.
Cydorhamphus marmoratus Garman, Bull. M. C. Z., 1875, 3, p. 276.
TelviatoUus aemaricus Boulenger, Cat. Batr. Sal. Brit, mus., 1882, p. 191.

Diagnosis. A small species with smooth skin; a strong supratympanic
fold; vomerine teeth present in two small or but moderately developed groups;
skin loose on the sides only and belly uniformly pale.

Habitat. Reported from about Lake Titicaca (type-locality) and
from other stations nearby; as from Vincocaya (Ganrian); Cuzco
(Cope, Proc. Amer. philos. soc, 1877, p. 39); Arequipa (Garman),
and Yura near Arequipa (Cope, Loc. cit.).

Type. x\cad. nat. sci. Phil., 11,400 from Lake Titicaca, Peru. ■

Description of Type. Size moderate; head broader than long, much nar-
rower than the body, its length contained in the total length of body 3.4 times;
snout rounded and flat without canthus rostralis; nostril nearer the orbit than

Fig. 7. — Telmatobius aemaricus (Cope). 1. Foot. 2. Hand. 3. Open mouth.

the labial border. Vomerine teeth barely visible in two small groups between
the choanae which are also small; tongue moderate in size, nearly round. In-
terorbital space 1.1 as broad as the length of the eye, the length of the snout
1.2 as long as the length of the eye; tympanum entirely concealed and covered

420 bulletin: museum of comparative zoology.

by a loose flap of skin. Digits free, stout, slightly dilated at the tips; the first
and second fingers equal in length and only a trifle shorter than the fourth; the
elbow extended forward reaches the posterior edge of the orbit. Toes fully
webbed, but the webs so fully notched that they appear only half webbed; a
free border of skin on the inner side of the tarsus and another on the outer side
of fourth toe; subarticular tubercles slightly developed; a distinct inner and a
low outer metatarsal tubercle; heels nearly in contact when the hind limbs
are folded at right angles to the axis of the body; the tibiotarsal articulation
reaches about two millimeters behind the posterior edge of the orbit when
the hind limbs are carried forward along the body. Skin smooth except for a
few low tubercles on the head and shoulders, the tubercles most abundant in
the occipital region; clusters of horny spines, characteristic ofthe male in the
breeding season, grouped on the chest and outer side of thumb; a well-de-
veloped supra tympanic fold which extends to the groin; two transverse folds
and a number of flat tubercles just posterior to the angle of the mouth; several
folds on the sides of the body; two loose flaps of skin extending the length of
the posterior side of the thighs.

Color above dark brown indistinctly mottled and spotted with darker brown,
color below uniform yellowish grey.

Dimc7isio7is.

Tip of snout' to vent 46 mm.

Tip of snout to posterior end of mandibular bone 16 "

Greatest width of head 15.5 "

Fore leg to tip of longest finger 28 "

Hind leg to tip of longest toe 64 "

Remarks. Cope says in the type description: — " Labial integument
largely free and overhanging the lower jaw." We have examined
carefully the five paratypes (Acad. nat. sci. Phil. 1,435-1,457, 16,177,
16,178) and it is certain that the labial integument of the type-speci-
men has been artificially torn away from the skull. The paratypes
are much better preserved than the type and the labial integument
of each one is normal. The labial integument is likewise attached
to the skull, normally, of course, in the ten specimens, nine from
Vincocaya and one from Arequipa (M. C. Z.).

Telmatobius jelskii (Peters).

Pseudobatrachus jelskii^ Peters, Monatsb. Berl. akad., 1873, p. 415.
Cyclorhamphus marmqratus Gtinther, P. Z. S. London, 1859, p. 89.
Telmatobius jelskii Boulenger, Cat. Batr. Sal. Brit, mus., 1882, p. 191.

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 421

Diagnosis. A medium-sized frog with smooth, somewhat loose skin; with a
strong supratympanic fold; vomerine teeth not greatly developed, in two small
groups; the tibiotarsal articulation reaching nearly to the eye and having a
dark sooty grey ventral surface except for the throat which is conspicuously
pale.

Habitat. Mountain streams of the high Andes of Central Peru.
Description. Adult M. C. Z. 4,790 from Tarma, Central Peru 1916;
John M. Boutwell.

Size moderate; head scarcely broader than long, about equal to width of
body, its length contained in total length of body 3^ times; snout declivous,
rounded, somewhat projecting, without canthus rostralis; nostril slightly
nearer orbit than tip of snout. Vopierine teeth in two tiny groups, very feebly
developed, situated directly between and nearly touching the choanae; each
aperture being much larger than a group of the teeth; tongue moderate,
slightly longer than broad. Interorbital space very slightly wider than
upper eyelid; the length of the snout one and one half times that of the
eye; tympanum hidden. Digits free, scarcely dilated at the tips, the first
finger equal to the second, much shorter than the fourth; the elbow extended
forward reaches the center of the eye. Toes extensively webbed, but webs
incised to appear but half developed; a distinct tarsal fold; subarticular
tubercles distinct, the inner metatarsal twice as long as the outer; heels just
touching when the hind limbs are folded at right angles to the axis of the
body; the tibiotarsal articulation reaching the posterior border of the eye
when the hind limb is carried forward along the body. Skin above glandular,
smooth or slightly spiny, a well-developed dermal lateral fold from orbit to
groin; skin on sides of body very loose and baggy; abdominal area ill-defined
by folds; skin of thighs loose and baggy also.

In the male described, there is a prominent area covered with fine asperities
on the chest as well as the usual digital asperities, which on the thumb are
extraordinarily developed.

Color uniform slate above, faintly spotted with darker; ventral surface
dirty yellow washed with dusky grey; a pale area beneath each thigh; throat
pale, unspotted.

Dimensions.

Tip of snout to vent 55 mm.

Tip of snout to angle of jaw 20 "

Greatest width of head 21 "

Fore leg to tip of longest finger 32 "

Hind leg to tip of longest toe 79 "

Remarks. This species, described by Peters, was obtained during
Count Jelski's journey in Peru, and the type-locality is the small

422 bulletin: museum of comparative zoology.

village of Acancocha near Vitoc. At our request Mr. John M.
Boutwell, a mining geologist at the Cerro de Pasco mines, kindly
made a special effort to secure a series of this little-known species.
He procured two lots of beautifully prepared examples from near
Tarma and Palca, two stations in the Oroya district and practically
topotypes.

Telmatobius montanus Philippi.

Telmatobius montanus Philippi, Supl. Batr. Chilenos Santiago, 1902, p. 47.
(?)Telmatobius aemaricus Werner, Zool. jahrb. Suppl., 1897, 4, p. 263.
(?) Telmatobius marmoratus Werner, Zool. jahrb. Suppl., 1897, 4, p. 263.

Diagnosis (extracted from original description). Indistinguishable from
T. laevis except for the narrower head and the toes which are fully webbed.

Habitat. A lake in the high Andes of the Province of Santiago,
Chile.

Remarks. Tdmatohius montanus is evidently the water-form of
T. laevis. It bears the same relation to T. laevis which T. culeus bears
to T. aemaricus. Philippi in describing T. laevis states that it comes
from a pasture, while he states clearly that T. montanus was found in a
lake. If Philippi's descriptions may be relied upon, it is very prob-
able that the same evolution with change of structure has taken place
in Chile as well as in the Andes of southern Peru, where there has-
been a change from a land to a completely aquatic mode of life.

Telmatobius laevis Philippii.
Telmatobius laevis Philippi, Supl. Batr. Chilenos, Santiago, 1902, p. 43.

Diagnosis (extracted from original description). No vomerine teeth;
choanae very large. Nostrils nearer the eye than the tip of the snout. Tym-
panum hidden, covered by undifferentiated skin. Toes one fourth webbed.
Skin entirely smooth, no glands on any part of the body. Color above black,
no trace of marldngs; ventral surface light grey, similarly without markings,
fingers somewhat lighter in color, especially at the point.

Habitat. Chile; range probably restricted to some of the pasture-
lands of the Andes. Philippi {Loc. cit., p. 44) states that the type-
specimens come from "Potrero," in other words from a pasture.

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 423

There are five towns known as Potrero in Chile. The specimens
probably came from some one of these localities rather than from
some wholly indefinitely located pasture.

Telmatobius marmoratus (Dumeril and Bibron).

C yclormnphus marmoratus Dumeril et Bibron, Erpet. gen., 1841, 8, p. 455.
Cyclorhainphus marmoratus Peters, Monatsb. Berl. akad., 1873, pi. 2, fig. 2,.

pi. 3, fig. 3.
Telmatobius marmoratus Boulenger, Cat. Batr. Sal. Brit, mus., 1882, p. 192.

Diagnosis (extracted from original description). Vomerine teeth in two
very small groups between the choanae. Toes half webbed. The tarso-
metatarsal joint reaches the tip of the snout when the hind leg is extended
forward. Skin perfectly smooth, neither glands nor tubercles on any part of
the body. Ground-color p,bove grey varying to brown; upper surface marbled
with black and generallj- irregularly covered with little white spots. Ventral
surface grey, with or without black marblings, the appendages darker than the
throat or abdomen.

Boulenger (1882, p. 192) adds in part to this description: — ^ Choanae
large. Tympanum small, hidden. The hind hmb carried forward along the
body, the tibiotarsal articulation does not reach the eye. A flat parotoid
gland frequently indistinct.

Habitat. Huasacona, a hacienda in the District of Asangaro, De-
partment of Puno, Peru.

Remarks. Dumeril and Bibron (1841, p. 455) in describing the
species say: —

"Cette espece est une decouverte faite au Chih par M. Pentland; le lieu oil
elle a ete trouvee par ce savant naturaliste se nomme Guasacona."

We have made a vain search for this locality in all the old maps and
statistical volumes at our disposal. There seems to have been no
locality by that name in either Chile or Bolivia. There is, however,
in Peru just one Guasacona, a hacienda of 221 (Resumen * * * habi-
tantes del Peru en 1876, 1878, p. 104) or 427 (Soldan, Dice, geogr.
estad. Peru, 1877, p. 427) inhabitants. Telmatobius marmoratus
has been recorded from Chile by Werner (1897, p. 263), but this is
most probably a case of misidentifi^cation. There is no good evidence
to show that any specimens of T. marmoratus besides the types have
ever been taken.

424

bulletin: museum of comparative zoology.

Telmatobius angustipes (Cope).

Cyclorhamphus angustipes Cope, Proc. Amer. philos. soc, 1877, 17, p. 38.
Telmatobius angustipes Boulenger, Cat. Batr. Sal. Brit, mus., 1882, p. 192.

Diagnosis. A rather small species, with smooth skin, no supratympanic
fold; vomerine teeth present and moderately developed; the tibio tarsal
articulation reaching the anterior border of the eye and in color uniform brown
above.

Habitat. The tjy'pe of this distinct and long-legged species formed a
part of the collection of Prof. James Orton's fruitful journey to Peru
in 1876-77. The single example which Cope described came from
Juliaca, a few miles west of Lake Titicaca, a desolate old town at an
altitude of 12,550 feet.

Type. Acad. nat. sci. Phil. 11,389 from Juliaca, Peru; altitude
12,550 ft.

Description of Type. Size small; head exactly as long as broad, slightly
narrower than body, its length contained in the total length a little more than
three times ; snout rounded, high but with the canthus rostralis obsolete, nostrils
slightly nearer the orbit than the labial border. Vomerine teeth in two small
groups between the choanae which are large and at an angle to each other,

Fig. 8. — Telmatobius angustipes (Cope). 1. Foot. 2. Hand. 3. Open mouth.

tongue longer than broad. Interorbital space 1.25 as broad as the length of
the eye ; the length of the snout 1 . 1 as long as the length of the eye ; tympanum
entirely concealed by the skin. Digits free, slender, without discs, the first
finger a trifle shorter than the second, much shorter than the fourth; the elbow
extended forward reaches nearly to the center of the orbit. Toes slender,
without discs, about one third webbed; subarticular tubercles not distinct;
only a slight indication of the inner metatarsal tubercle; heels slightly over-

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 425

lapping when the hind limbs are folded at right angles to the axis of the body;
the tibiotarsal articulation reaches a little beyond the orbit when the hind
limbs are carried fon\'ard along the body. Skin entirely smooth. No folds
or creases on the body.

Color above uniform dark brown; hght yellowish brown below.

Dimensions.

Tip of snout to vent 37 mm.

Tip of snout to posterior end of mandibular bone 13.5 "

Greatest width of head 12.5 "

Fore leg to tip of longest finger 24 "

Hind leg to tip of longest toe 58 "

Remarks. Telmatohius angustipes is closely related to T. marmora-
tus, but differs from it in several characters, the most important of
which is the greater leg length. It is only in a country of such strik-
ing physiographic barriers as Peru that two species so closely related
may occur so near each other.

BUFONIDAE.

BuFO MARiNis (Linne).

This extremely widespread species was met with at Palambla (8
specimens), Perico (6 specimens) and at Chumayo (5 specimens) while
a large series of tadpoles and very young were secured at Bellavista.

BuFO SPiNULOSUS Wiegmann.

This wide ranging Andean toad was found at Huancabamba where
three large adults were caught and at Chumayo where three others
were taken. These bear close comparison with specimens from
southern Peru and the mountains of Chile and Argentina.

BuFO TYPHONius (Linne).

This form so characteristic of the tropical lowlands of South Amer-
ica was found abundantly at Perico, Bellavista, and Querocotilla
while four were found in the rain forest at Tabaconas.

426 bulletin: museum of comparative zoology.

Hylidae.

Gastrotheca monticola, sp. nov.

Diagnosis. Closely related to G. marsupiatum from which it may be dis-
tinguished by the following characters: — Vomerine teeth in two straight con-
fluent series on a level with the hinder edge of the choanae. Snout a little
longer than the diameter of the eye. Loreal region slightly concave. Inter-
orbital space once and a third the diameter of the eye, strongly concave. Toes
two thirds webbed. The hind leg being carried forward along the body the
tibiotarsal articulation reaches between the eye and the nostril. Skin smooth
or finely granular above, generallj' granular on the head and sides. Size large,
the pouched female at least sixty-one millimeters from snout to vent, generally
sixty-eight millimeters, coloration somewhat different from G. rnarsupiatum;
female with irregular dark blotches on the back, never forming two sym-
metrical stripes as in that species.

Range. Only known from the low central Andean range of northern
Peru from Huancabamba south to Querocotilla.

Tyjje. M. C. Z. 5,290 from Huancabamba, northwestern Peru; 15
August, 1916, G. K. Noble.

Description of Type (adult female with empty pouch). Size large; head
about as broad as the body, much broader than long; snout blunt forming
a semicircle with distinct canthus rostralis and concave loreal region; orbital
diameter equals the distance of the eye from the nostril, which is very near the
end of the snout; interorbital space about twice as broad as an upper eyelid.
Tongue large oval, slightly emarginate behind. Vomerine teeth in two
ovoid groups, touching each other and forming a straight line between the
posterior halves of the choanae. Tympanum three fifths the size of the orbit.
Fingers with a very slight rudiment of a web; the first finger equal in length
to the second; toes two thirds webbed, e. g. two inner toes webbed to base
of penultimate phalanx, third to middle of penultimate, fourth about one
third the length of the antepenultimate and fifth almost to the end of the
penultimate; discs distinct, about one third of their width broader than the
penultimate phalanx, distinctly narrower than the width of the tympanum;
subarticular tubercles well developed; a large inner metatarsal tubercle and a
distinct fold along the inner side of the tarsus. Tibiotarsal articulation
reaches the nostril, or not quite so far. Skin finely granular on the back,
coarsely granular on the sides and very coarsely granular on the ventral
surface; a slight indication of a dorsolateral fold.

Color in alcohol generally bluish grey above; a dark greyish spot between
the eyes; a broader one in the scapular, and a narrower one in the iliac region.
(In life the ground-tone was brilliant green, and the pattern was dark brown) ,

BARBOUR AND NOBLE: AMPHIBIANS FROM PERU. 427

a dark stripe along the canthus rostralis and dorsolateral fold; sides of the
bodj' heavily marbled with black and white; lips white; appendages barred
with greyish brown; posterior surfaces of the thighs and under sides of the
legs mottled with dark brown; ventral surface of the head and body white,
immaculate.

Dimensions.

Distance from snout to the vent 63 mm.

Greatest width of head 23 "

Distance from axilla to tip of largest finger 40 "

Distance from groin to tip of largest toe 101 "

Remarks. Gastrotheca monticola is readily distinguishable from all
the specimens of G. viarsupiatum examined by its much larger size.
The several pouched females of G. marsupiatum in the M. C. Z. from
Ecuador vary from 43 to 48 millimeters in length (snout to vent),
while none of the ten pouched females of G. monticola in our series are
less than 61 millimeters in length.

Unlike G. boliviana and perhaps other species of Gastrotheca, this
species exhibits a well-marked sexual dimorphism. The males in life
were alwaj^s some shade of tan and were heavily blotched with brown.
These markings often formed a ) (-shaped figure in the pectoral region.
The females were always green and were less heavily marked. Some-
times the females were uniformly green above except for the eye-stripe
and a few dark markings on the sides of the body.

EXPLANATION OF THE PLATES.

PLATE 1

Barbour & Noble. — Amphibians from Pern.

PLATE 1.

Sminthillus limbatus (Cope).

Fig. 1. — Dorsal view.

Fig. 2. — Ventral view.

Specimen cleared after Schultze's method by Thomas Barbour in
1913.

BULL. MUS. COMP. 200L.

Barbour and Noble. Peruvian Amphibians. Plate 1

PLATE 2.

Barbour & Noble. — Amphibians from Peru.

PLATE 2.

Fig. 1. — Gastrotheca monticola Barbour and Noble.

Female with pouch containing an advanced embryo. A tadpole
may be seen at the orifice.

Fig. 2. — Gastrotheca monticola Barbour and Noble.

Male. The male is always more striped than the adult female.

Fig. 3. — Humid subtropical zone near Charape.

BULL. MUS. COMP. 200L.

Barbour and Noble. Peruvian Amphibians. Plate 2

PLATE 3.

Bahboub & Noble. — \mphibians from Peru.

PLATE 3.

Fig. 1. — Arid subtropical zone near Querocotilla.
Fig. 2. — Interior arid plain near Perico.

BULL. MUS. COMP. ZOOL.

Barbour and Noble. Peruvian Amphibians. Plate 3

|r,.:.

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE.

Vol. LXIII. No. 9.

DOGS OF THE AMERICAN ABORIGINE^

By Glover M. Allen.

With Twelve Plates.

CAMBRIDGE, MASS., U. S. A.:

PRINTED FOR THE MUSEUM.

March, 1920.

No. 9. — Dogs of the American Ahoricjincs.
By Glover M. Allen.

CONTENTS.

Intixxluction .....

Acknowledgements

Origin of the Domestic Dog .

Origin of Amei'ican Dogs

Breeds of American Aboriginal Dogs

Eskimo Dog ....

Plains-Indian Dog .

8ioux Dog ....

Long-haired Pueblo Dog

Larger or Common Indian Dog

Klamath-Indian Dog

Shoit-Iegged Indian Dog

Clallam-Indian Dog

Inca Dog ....

Long-haired Inca Dog

Patagonian Dog

Mexican Hairless Dog

Small Indian Dog or Techichi

Hare- Indian Dog

Fuegian Dog ....

Short-nosed Indian Dog (Pachycyon)

Peruvian Pug-nosed Dog
Summary .....
Bibliography ....

Explanation of the Plates

Page.
4.31
432
432
439
440
442
449
4,55
456
457
463
464
469
472
475
476
478
481
491
492
495
500
.503
504

Introduction.

When Columbus, in 1492, made his discovery of land in the Western
Hemisphere, he found it already peopled by a race of men who are
considered by modern ethnologists to be of Asiatic origin, and probal)ly
of an anticjuity dating back not many thousands of years. Yet these
aboriginal peoples were considerably diversified as to appearance,
language, and customs. In South America, the Incas had domesti-
cated animals, llamas and alpacas, whose wild progenitors are the last

432 bulletin: museum of comparativk zoology.

remnant of the once diverse phylum of American camels. There is
no good evidence, however, that the horse which survived in North
America till late Pleistocene times M^as ever known to the aborigines
until its reintroduction b;s' Europeans. Dogs they had, nevertheless,
universally and in some variety. Yet at this late date it is hardly
possible to define the various breeds or variations with any exactness
or to throw much light on the question of their ultimate origin. An
attempt is made here to gather what information the earlier travellers
recorded as to the appearance of the dogs of the American aborigines,
and so far as may be, to characterize the various breeds that can be
distinguished.

A bibliography is added giving the more important papers on the
origin of the dog, and on prehistoric dogs of the Old World, as well as
jeferences to the aboriginal dogs of America.

Acknowledgements.

For the opportunity of studying dog-remains from various parts
of the New World, I would express my obligation to the Musemn
of Comparative Zoology; to Messrs. C. C. Willoughby and S. J.
Guernsey of the Peabody Museum; to Mr. G. S. Miller, Jr., of the
U. S. National Museum; Prof. F. B. Loomis of Amherst College;
Prof. W. K. Moorehead, of Andover Academy; and Messrs. A. L.
Kroeber and E. W. Gifford of the Museum of Anthropology' of the
University of California.

For interesting photographs of dogs, thanks are gratefully extended
to Messrs. Ernest Harold Baynes, W. B. Cabot, C. T. Currelly,
W. C. Farrabee, S. J. Guernsey, the Royal Ontario Museum of Arch-
aeology, and the American Genetic Association.

Origin of the Domestic Dog.

The problem of discovering the wild ancestor of the Domestic Dog
has engrossed the attention of naturalists from the time of Buffon to
the present. Basing their opinion on general external resemblances,
the early systematists, Giildenstadt and Pallas, favored the Indian
Jackal as the primitive stock whence the European dogs were derived.
In this course they have been followed by many later writers, but more
exact studies (Miller, 1912) show that the teeth of the Jackal may be

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 433

distinguished l)y man\- minor characters (such as the broadh' con-
tinuous outer cingukun on m- and m^) from those of the Wolf and Dog.
Gidley (1913) has iUustrated more fully some of the distinguishing
tooth-characters of several cani^ds, including fox, wolf, and coyote,
and has grouped them into a key, from which it is seen that domestic
dogs and wolves are essentially alike in the cusp-characters and pro-
portions of their teeth, and differ from coyotes and foxes in a\erage
characters which though slight, are appreciable on direct comparison.
Miller (1912, p. 313) concludes that in a series of dog-skulls "repre-
senting such different breeds as the pug, fox-terrier, bloodhound,
mastiff, ancient Egyptian, ancient Peruvian, Eskimo (Greenland and
Alaska) and American Indian, the teeth are strictly of the Avolf type";
and this assertion I can fully endorse from a study of these and other
breeds. Nevertheless, though the Wolf and the Domestic Dog are
closely related, it does not follow that the latter is directly derived
from the former, though even as lately as 1911, Trouessart has upheld
the view first put forth by Jeitteles (1877), that the Indian Wolf
{Cams pallipcs) might be the ultimate source of certain breeds of the
Dog. Studer (1906) suggests some large Dingo-like type as the lost
ancestor; while Noack (1907) supposes that the original stock may
have been identical with a small Chinese Wolf of which he possessed
two specimens from Tchili, regarded as like the Dingo in color. Xeh-
ring (1887) suggests that a small Japanese Wolf (C. japonicus) is the
living ancestor of the Japanese Street-dog. The Dingo itself is of
doubtful origin, and though probably a relatively recent arrival in
Australia, may have been brought at the time the Continent was first
peopled by man. Kreft't (1866) believes he has identified its "first
molar tooth. . . .with other fossil remains in the breccia of the Welling-
ton caves," while McCoy (1862) has "identified its bones mingled
with those of recent and extinct animals all in one state of preserva-
tion in the bone-caverns recently opened beneath the basalt flows at
Mount Macedon." In New Zealand, domestic dog-remains of a
different breed are found associated with those of the extinct giant
rails in the kitchen-middens And j:)resumably came with the Maoris
(Hutton, 1898).

The older naturalists maintained the view that cross fertility was a
test of specific identity, and recorded many cases in support of the
contention that the Dog was fertile with Wolf and Jackal, and that
hence it was of such mixed ancestry. Thus, Hunter (1787) recorded
the fertile cross between a male Dog and a female of the Wolf and of
the Jackal, whence he concludetl that all were of one species. A more

434 BULLETIX: MTSKI'M OF COMPARATIVE ZOOLOGY.

recent investigator, (Kiihn, 1SS7) records the fertilitv' of Dog-Jackal
hybrids when crossed infer .sc or l)ack crossed. In thiscase a female
Finnish Bird-dog was bred to a captive Indian Jackal {Canis- aureus
iudiruft), producing three litters of four each. All the young were
much alike in appearance reseml)ling the Jackal, but were somewhat
darker in color. One of the hybrids l)red to a Sil)erian Dog produced
seven young. Two other of the original hybrids were paired together,
and produced a litter of three young after a period of sixty days'
gestation — the normal time for a dog. These young were darker
than their parents, with a wash of golden along the sides and on the
head, recalling the Jackal's color. Unfortunately no careful study of
the cranial and dental characters in the hybrids was made.

The crossing of Wolf and Dog has been frequently accomplished in
captivity (Hunter, 17S7, 1789). An instance of the fertile crossing
of a Siberian Sledge-flog with a female Dingo from Australia is re-
corded b\- Eitfe (1909). The North American Indians and the the
Eskimo are accredited with tethering female dogs in heat at a distance
from camps to obtain crosses with wild wolves, which though usually
highly hostile to dogs, will at such times, it is said, hyl)ridize. Ac-
cording to Cones (1873) and others, similar methods were used by the
American Indians of the Plains to obtain crosses with wild coyotes.
Yet the evidence is not altogether convincing that such cross-breeding
was very general, or that it has modified the nati\'e dogs in any way.
It is noteworthy that the American Indian is not given to the domesti-
cation of Wolf or Coyote puppies as might be expected if either were
the prototype of his Dogs. Nevertheless (^oues (1873) and Packard
(1885) on the groimd of general external appearance ha^•e held that
the common Indian Dog of North America was merely a tamed
Coyote; and their view has gained wide credence. It may l)e con-
fidently stated, howe\er, from a study of skulls and teeth, that this is
not at all the case. Packarrl was perhaps influenced by Cope's
(1883, p. 242) statement that "many of the domesticated dogs have
been derived "from the Wolf and the Coyote, as found in the Pliocene
deposits of the Repuljlican River formations. The x\merican Indian
dogs, however, are true domestic dogs in skull-characters, and show
no e\"idence of deri\-ation from coyotes.

Crosses between domestic dogs and foxes have l)een less commonly
reported, and even these reports seem to lack proper substantiation
in most cases. B. Ross (1861) explicitly states that the dogs of the
northern Indians could not be induced to cross with capti\e foxes.
A supposed case is given by Toni (1897) of a natural hybrid, but its
ancestry as in one or two other cases, was merely conjectural.

ALLEN: DOGS OF THK AMERICAN ABORIGINES. 485

While some naturalists have thus sought to derive the Domestic Dog
from Wolf, Jackal, Coyote, or Fox, or from a mixture of two or three
of these, others have maintained that it is quite as well entitled to be
considered a distinct species with its various artificial breeds. Buffon
was one of the first to support this N'iew. Pictet (ISoo, 1, p. 203-210)
believed that do^remains from ca\e-(leposits in Europe prol)ably
represented the wild ancestor of domestic dogs, and to this wild
species he gave the name Cams fainUiaris fossili.s. In this he was
followed by Bourguignat (lS7o) who regarded the Prehistoric Dog as
a species, related to the Wolf but coexistent with it in a wild state.
He applied to it tiie name Cauls fcrus, and concluded from the relative
scarcity of its remains in the earlier strata of human culture, that it
was at first seldom domesticated by the early cave-men. Remains of
Pliocene canids from central France have been suggested by Boule
(1889) as representing the progenitors of the Domestic Dog.

Although the recent and more exact studies of Miller (1912, ]). 313)
and Gidley (1913, p. 99) have shown that the Domestic Dog nuiy be
distinguished l)y dental characters from Coyote, Jackal, and Fox, its
close relationship to the wolves is shown, as they point out, l)y the
shorter and narrower heel of the lower carnassial in proportion to the
length and width of the remaining part, the general bluntness and
plumpness of the premolar and molar teeth and their cusps, as well
as by the shorter and blunter canines. , Other less constant but
average distinctions are tai)ulated by the latter author. A noticeable
character of the lower tooth-row in Wolf and Dog may also be men-
tioned, namely, its distinctly outward bend at the junction of the
molar and premolar series, whereas in the Coyote and the Jackal, the
axis of the tooth-row is much more nearly a straight line. The
presence of a minute second posterior cusp in addition to the cingu-
lum in the fourth lower premolar is characteristic of Jackal and Coyote.

The relationship of the Domestic Dog having thus been found to
be wholly with the Wolf, and not with Jackal, or Coyote, it remains
for future investigation to show what wolf-like ancestor was its wild
progenitor. This, however, lies outside the scope of the present
paper. Yet it may l)e said that no evidence has hitherto been ad-
duced that clearly indicates the origin of the Dog from any of th(>
large wolves of circumboreal distribution. In general the skull of
the Dog is at once distinguished from that of the Wolf, apart from its
usually smaller size, by the higher forehead of the former. That this,
however, is due to greater de^■elopment of the cerebrum through
domestication has been suggested by Hammeran (lS9o), notwith-

436 bulletin: museum of comparative zoology.

standing that domestication in case of most animals seems rather to
have a stultifying effect. A more diagnostic character is found in the
size of the teeth, Avhich e^•en in the largest breeds of dogs are con-
siderably smaller than in the wolves. A fact of probable significance
is that in wolves as in the less modified breeds of dogs, e. g., the
American Indian dogs, the free posterior border of the palate ends
about on a line passing transversely through the middle of the last
molar. In the large breeds of European dogs a transverse line at the
hinder margin of the palate usually falls considerably behind the last
molar, indicating probably that the teeth have retained more nearly
their original size relations than have the maxillar,^' and other bones.
A like condition is seen also in dogs in which the teeth are abnormally
reduced in size, due probably, as in case of the Chinese Chow Dog, to
a diet of soft foods as rice and fish through many generations. These
facts tend to indicate that the Dog and the large Wolf are really
distinct species, and that the wild progenitor of the Dog was a small
Wolf of a species distinct from the large wolves of circumboreal dis-
tribution. It is natural to look to Asia for this unknown ancestor,
and it would be valuable if the studies of Noack and Nehring as to
the small wolves of Tchili and Japan might be more fully confirmed.
Jentink (1S97) suggests the Wild Dog of Java as a representative of
the original stock whence the Domestic Dog sprang.

Attention should here be called to the possible effect of domestica-
tion in reducing the size and proportions of the Wolf. Apparently
the only in\'estigator to compare the skulls of wolves born in captivity
with those of wild indi\iduals is Wolfgramm (1894), who states that
the skulls of the capti\-e-born wolves are smaller in all proportions,
broader and higher, with less developed muscle-crests. The snout
is so shortened that pm'^ is forced to assume a transverse position,
the lower premolars are imbricate, while in size the carnas.sial as well
as the other teeth are said to be slightl;\- reduced. Wolfgramm con-
cludes that this i? stifficient proof that the Dog is derived from the
European Wolf, and that its smaller size is a direct resiilt of its do-
mestication. The facts, however, do not warrant such a conclusion.
The reduced size of the skull and the crowding of the teeth in captive-
born wohes are probably a result of improper nutrition during growth
and lack of exercise under confinement, conditions wholly different
from the free life of a dog imder domestication. The crowding of
the premolars is quite as abnormal for a dog as for a wolf, and occurs
through failure of the maxillar,\- bones to attain their proper growth,
while the teeth themselves attain their size independently.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 437

While some authors have considered that modern dogs are poly-
phyletic, and would trace the ancestry of the larger breeds to wolves
and of the smaller to foxes (Woldrich, 1886a, even suggests the Fen-
nee!), it seems more reasonable to derive them all from a medium-
sized dog through selective breeding. Nevertheless it is possible to
divide modern breeds into some four to six groups, based mainly on
size and minor external characters as erect or lop-ears, drooping or
curled-up tail, etc. Cuvier (1808) believed that the French Sheep-
dog approached the wild prototype most nearly of all domestic
breeds, and considered the Australian Dingo as the most primitive
true dog. The characters considered primitive are chiefly the medium
size, the erect, wolf-like ears, imshortened snout, drooping and
moderately haired tail, and low forehead. The ability to bark is
often considered an acquired trait; and the more primitive dogs,
such as the Eskimo, howl like wolves more than they bark.

Historic evidence as to the ancestry of the Dog does not carry the
matter far enough. The Egyptians had dogs as far back as the records
go — certainly four to five thousand years before the Christian era.
The same is apparently true of the Chinese, Avhose history goes back
nearly as far. Lertet and Gaillard (1909) recognize four breeds of
dogs among the mummified remains from Assiout. Fitzinger (1866)
has summarized the ancient history of dogs known from the earliest
writings of Rome, (ireece, Assyria, and Egypt. Yet it is clear that
at the dawn of history, the nations of Europe, Asia, and North Africa
had dogs of several breeds, more or less characteristic of each people.
Thus the Greyhound type seems especially prevalent in Egypt and is
to this day associated with the desert-loving races of Persia and
northern Africa.

European archaeologists have made many disco\'eries of dog-re-
mains in association with bones and implements of prehistoric man,
particularly in the caves and old Lake-dwellings of southern P^urope.
Hitherto at least eleven different Latin names have been applied to
as many supposedly distinct prehistoric dogs of P^urope. Anutschin
(1881) announced the discovery of the first dog-remains to be found in
Russia. Parts of fourteen dog-skeletons were found in building the
Ladoga Canal, and represent two types which he names respectively
Cams familiaris palustriti ladogcusis, and C. f. inostranzcivii. He con-
siders these to be of the Stone Age, and that the former is closely allied
to the Siberian and Northwest American Sledge-dogs — (Eskimo).
The latter he thinks Aery similar to the C. mafri.s-optiiiiac, a deer-
hound-like type, from the Bronze Age, or even earlier (Neolithic,

438 BT'LLETIX: Ml'SETM OF COMPARATIVK ZO(")LOGY.

according to Xehring, 1883). Dog-remains, associated with a human
skeleton and palaeoHthic implements, were described by Studer (1906)
as Canis pmdiatini, and were discovered while digging a street near
Gute Bologo'ie in Russia. This was as large as a medium-sized Sheep-
dog and is believed l)y tliis author to be the fore-runner of (\ nifrr-
mrdhis of the Bronze Age, which is possibly a hound.

In the Swiss Lake-dwellings occur skulls of a smaller type of dog
named by Riitimeyer Cmiis pahi.stri.s-, a l)reed characteristic of the
later Neolithic and the Bronze Ages, in Europe, o,000 to 7,000 years
ago. Another Neolithic Dog of small size fskull length, ]r)8 mm.) is
described by Hue (1900) from ('lair\aux. Jura, as ('(uii.s Ic itiirri, while
still another of dwarf proportions, f. iiiikii, is considere<l by vStuder
(1906) as a fore-runner of C. jjal n.s-fris. The same author (Studer,
1901) sees much resemblance between skulls of C. pcdustris and those
of Chow and Spitz. rndoul)tedly the Chow is a rather ancient type,
in numy ways recalling the Eskimo Dog in its erect short ears,
broad muzzle, small eyes, bushy mane, and curled-up tail carried
stiffly over the hip. Measurements of skulls of Chows given by
Stufler are slightly larger than those of C. paluMri.'i.

No less than four breeds of dogs are recognized b\' Strobel (ISSO) in
human cultiu'e layers transitional from the Neolithic to the Bronze
Age in Emilia, Italy. One is the small C. palu.s'fris wide-spread in the
Stone Age of Europe; the second is C. htfcrDicflins, a larger dog sup-
posed to be a hound; the third is the larger C. )iiafn'.s'-optimar, re-
garded by Studer (1901) as of the Collie and Sheep-dog (Wolf-dog)
type, while the fourth is a Dog smaller than palu.ifris, and believed to
be of a distinct breed which Strobel names C. .spaldti. Remains of
the first three of these breeds are recognized by Woldrich (1898) from
culture layers of middle Neolithic times in caverns of Bohemia.

From these brief accounts of discoveries of prehistoric dogs it is
clear that at a very early period of human culture there were at least
two or three types under domestication in Europe. It need not be
supposed, as some authors have done, that these types are of local
origin. Europe, as a peninsula of Asia, probably recei\'ed its dogs as
well as its human population in part at least from the East. Possibly
then, as now, certain breeds of dogs were characteristic of different
invading tribes.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 439

Origin of American Dogs.

Very little attention has been paid to the dogs of tlie American
Aborigines. At the present day it is prol)ably too hite to find pure-
bred examples of most of the local varieties that formerly occurred.
Barton (1805) was about the only American naturalist to give much
thought to the matter, but the few notes he collected were taken
mostly at second-hand and were rather indefinite. Coues, Cope, and
Packard, as well as many writers following them, considered that the
domestic dogs of America must have been derived from the Coyote,
or from some other inrligenous species of North or South America.
Cope was the only one who made an examination of the teeth. In a
fragment of a lower jaw from Florida, Cope (1893) made particular
note of the absence of the first premolar and remarked on the large
size of the metaconifl and the entoconid of the lower carnassial. It
is true that in a large percentage of American nati\-e <logs the first
premolar is al)sent from the lower jaw. A similar anomaly is occasion-
ally seen in wolves and in European dogs, but is rare. It is usually
considered that the first premolar in dogs is without a milk prede-
cessor, but though tliis is often true, it is not always the case. A
jaw of a very young dog in the ^Museum collection, shows very small
milk-teeth capping the permanent first premolars which are nearly
erupted. A similar case is reported by Lataste (1888). The entire
suppression of the first premolar, particularly in the lower jaw, in
a large percentage of American dogs, is possibly a retention of the
usual early condition, in which there is no first milk premolar.

The important paper of Loomis and Young (1912) and the reports
of Nehring on dogs from ancient Peruvian burials comprise most of
the work that has been done in the comparative dental and osteologi-
cal study of American dogs. There are, however, brief notices of the
discovery of prehistoric dog-ramains and early accounts of certain
native dogs by tra\ellers, the more important of which are included
in the Bibliography (p. 504-017). Miller (1912) seems to have been
the first to show that the teeth of American aboriginal dogs are those
of true dogs rather than of coyotes or wolves. This I haxe ^•erified
from a considerable mass of material from North America and Peru,
so that there can be no question but that the domestic dogs of both
Old and New Worlds are closely related and of common ancestrw
It follows that instead of having domesticated various dog- or fox-like
species of the Amei-ican continents, the peoples of the New World

440 bulletin: museum of comparative zoology.

must have brouj^ht their dogs with them, presumably from Asia, and
this probably at a culture stage prior to the domestication of other
animals, at least in the North, since no other domestic animal is com-
mon to the peoples of both hemispheres. The Asiatic origin of Ameri-
can dogs has previously l)een suggested by Mercer (1897, p. 12G) and
Wissler (1917).

The probability therefore is, that the Domestic Dog originated in
Asia and was carried by primitive man both east and west into all
parts of the inhabited world. That this migration began in late
Pleistocene times seems highly probable.

In the Western Hemisphere three types of dogs may in a very
general way be distingui.shed : — (1) the large wolf-like Eskimo Dog
of the Arctic countries, strong, powerfully built, with broad muzzle,
erect ears, and large bushy tail curled forward o\er the hip; (2) a
smaller type, ^•arying more or less in size and proportions, with erect
ears but a drooping tail; and (3) a much smaller type, the size of a
terrier, heavy of bone, usually with shortened rostrum as seen among
the tribes of the Southwest or again, apparently more slender both in
limb and skull as in southern Mexico or parts of South America.
South of the Eskimo country, the two latter types of dogs are char-
acteristic, and seem to have occurred together over much of their
range, so that travellers often mentioned a "wolf-like" and a "fox-
like " dog among the Indians of both North and South America.
In this connection, it is interesting to recall Kohler's (1896) statement
that in eastern Asia, between the provinces of Gansing and Ussuri,
the Chinese have small fox -like dogs, a comparison of which with the
small American dogs would be of interest. The smaller American
dogs of the slender type (Techichi) seem not xery different from the
Old "World C. palustris, and may be not remotely related. The more
heavily built small dogs with shortened faces and shorter, stouter
limb-bones, are perhaps deri\'ed from the more slender t\'pe, and
possibly owe certain of their peculiarities to cross-breeding with the
larger dogs, though this is at present wholly conjectural.

Breeds of American Aboriginal Dogs.

While in a \ery general way it may be said, that excluding the
Eskimo Dog, the American Indians had domestic dogs of two chief
types, a larger and a smaller, there were apparently sundry local breeds
of these, probably conforming in distribution with the general areas

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 441

occupied by the groups of tribes amongst which they were fotuul.
In the following pages tin attempt is made to define such of these
breeds as seem to be indicated by the fragmentary accounts of
travellers as well as l)y the study of what skeletal remains have been
available. No doul)t the number of breeds recognized is subject to
revision, for it has been found difficult to determine with any approach
to certainty in some cases, what external and skeletal characters are
to be associated, and in how far certain supposed breeds are mongrel
or relatively pure. Again, the skeletal characters may frequently fail
to give any clue to external traits that would be distinctive. More-
over, while the term " breed" is applied to these locally distinct forms
of dogs, it is not assmned that the American nati\'es made any con-
scious effort to change or keep constant the traits of their dogs;
possibly some of the variations are merely the result of a certain
mongrel mating, going on quite independent of human intent, so that,
as in case of the Peruvian Pug-nosed Dog, the variation cropped out
only occasionally and may or may not have been purposely preserved.
Nomenclature. — The bestowal of Latin names upon the different
breeds of dogs recognized has here been purposely avoided, as it
seems unwn'se to extend to such artificial variations the systematic
recognition accorded natural species and subspecies. Nevertheless,
Latin names or Greek letters have been used by other writers to indi-
cate domestic breeds, and such names have been applied in many
ways : — as trinomials, quadrinomials, or quinquenomials ; some-
times separated from the binomial, Cards familiaris, by a comma or
the abbreviation "var.," or othei"wise used in such a way as to cause
doubt as to their technical standing in systematic nomenclature.
Some names of dogs have been erected in a strictlv binomial fashion
and if accorded standing, conflict with other names. Thus Riiti-
meyer's Canis jjalustris (1863) of the Lake-dwellings is preoccupied
by von Meyer's Canis {= Gqlecynus) palustris (1843). The name
Cams mexicanus currently used for the Mexican Wolf pro\es to apply
to the Mexican Hairless Dog only. Hodgson's Canis lanigcr (1845)
for a Thibetan Wolf is preoccupied by Hamilton Smith's Canis lanigcr
(1840) for the Nootka Sound Dog. Other cases might be added.
The practice of using standard English (or vulgar) names for all arti-
ficial breeds is therefore to be recommended. With the descriptions
following, a list of Latin names applied by pre\ious writers is gi\en
under each breed.

442 bulletin: museum of comparative Zor)LOGY.

EsKLMo Dog.
Plate 1, fis. 1.

1817. C'anis familiaris tiihiricus groenlandicus Walther, Hund, p. 27 {fide
Fitzinger; not Cards groeidmuiicv.s Bechstein, 1799, q. e. Alopex).

1820. C.f. var. n. horealis Desniarest, Mamm., 1, p. 194.

1840. Cnnis horealis Hainilton Smith, Jardine's Nat. library. Mammalia,
10, p. 127, pi. 2.

Characters. — Size large, appearance wolf-like, but with less ol)lique
eyes, less attenuated muzzle, and more elevated forehead; tall usually
carried curled forward over the hip: teeth much smaller than those of
the Wolf. Pelage thick, with a shorter under fur o\'erlai<l with longer
hair which on the shoulders may be as much as eight inches long; tail
bushy. Color whitish, more or less clouded on the back, with dusky,
or varying to black, or black and white, or rarely tan and white.

Di.sfribufion. — The Eskimo Dog was originally found in Arctic
America coextensively with the Eskimo tribes from the barrens of
Alaska to Labrador, chiefl\- along the coast. In the east it was
probably at its southern limit on the east coast of New^foundland, and
thence ranged northward, accompanying its Eskimo masters, to Smith
Sound, Greenland. In Greenland it formerly was found along the
west coast southward, with the natives, but the present-day sledge-
dogs of the Danish settlements are probably largely jnongrel, through
interbreeding ^\■ith dogs introduced from Europe (Brown, 1875); and
the same is true of those in Alaska and southern Lal)rador.

External Measurements. — An I^skimo Dog brought back by Parry,
on his first \oyage, is figured by Children (1827) who gives its dimen-
sions as follows : —

Length, occiput to i-oot of tail 28 inches about 71 cm.

" " end of nose 11 " " 28 "

of tail (about) 18 " " 45.7 "

Total length (therefore about) 57 " " 145

Length of oar 3 " " ' 7.7

Ej'^es to point of nose 4 " " 10

Standing height at shoulder 24 " " 61

• These figures do not indicate a very large animal. The very thick
coat, especially on the shouklers, gi\es an increased appearance of size
not well borne out by skeletal measurements. It should be kept in
mind, that since the advent of Europeans, much attention has been

ALT.KX: DOGS OF THE AMERICAN ABORIGIXES. 44-'^

uivcn to increasing the size tmd strength of these northern dogs for
draught purposes. It is likely that the large wolf-like Eskimo Dogs
now common in the North, are considerably different from the original
stock found by the early Arctic explorers.

Figures. Children, J. G. Zool. journ., 1S27, 3. ])l. 1. From Parry's first

voyage.
Audubon, ,T. .J. and Bachnian, .T. Quach'upeds of Xorth America, 1848, 3, j)l.

113. Zoological Gardens, London.
Smith, C. Hamilton. Jardine's Nat. librar^^ Mammalia, 1S4(), 10, j)!. 2.

Prince's Street Gardens, Edinburgh.

Cranial Character^!. — Among the various skulls of so-called Eskimo
Dogs examined, there is more or less disparity of size. This is no
doubt an indication of the extensive crossing with European dogs
that has been carried on for a long period with a \'iew to improving
the speed and strength for which this dog is useful. Skulls from
eastern Kamtschatka are small, others frojn Alaska and Mackenzie
are of superior size. It is therefore difficult at the outset to determine
what the original Eskimo Dog of Xorth America was really like. It
is notable, lioAvever, that the teeth, even of the largest skulls are not
nuich larger than those of medium-sized skidls, while in no case do they
approach the magnitude of the Wolf's teeth. It would be of the
utmost interest, in this connection, to compare the teeth of a known
hybrid between the Eskimo Dog and a Wolf. Yet in spite of tiie fre-
(luency with which this cross is said to occiu", there seem to be few
skulls available. Windle and Hujnphreys (1S90, p. 9) give the r;itios
of dift'erent parts of such a skull to the basicranial axis.

For lack of a more authentic standard, I lla^'e taken as typical of
the P'skimo Dog, portions of a skull (M. ('. Z. 10, .537 10,539) ex-
humed by Dr. M. P. Porsild from an old village site at Sermermiut,
west Greenland. While not of great size, this skull is notable for its
broad palate, rather prominent trough-like depression between the
frontals, and the high strong sagittal crest, yet is the surface of the
brain-case comparatively smooth. Nearly similar is the skull of an
Eskimo Dog from Hebron, Labrador, collected in 1897. Its wide
palate and stout teeth are particularly noticeable as well as its strongly
developed crests and broad forehead.

444

bulletin: museum of comparative zoology.

M. C. Z.

m. c. z.

U.S.N. M.

Measurements of the Skulls

10,538

7.406

83,869

Greenland

Labrador

Baffin Land

Upper tooth-row, alveolus of i^ to m-

105

" c to 7«2

" pi to m«

" p' to m2

" //(' to m'

19 5

Length of carnassial, p*

19.5

Width of palate outside m^

Palatal length, alveolus of ?' to median edge

Lower jaw, alveolus of ii to tn-.^

105

— ^-

" c to w.

" Pi to m.

—

" Pi to ///.

" p, to 7n,

—

" Pi torn,

—

"mi torn,

—

Length of carnassial, nii

23.6

—

Width across postorbital processes

64 —

" " zygomata

125

. —

—

" " occipital condyles

NathusiuvS (1874) reports on ten skulls found near old Eskimo huts
in Jackson and Sabine Islands, Greenland. The largest of these had a
basal length of 189 mm., the smallest 175 mm. In skull U. S. N. M,
83,869 the basal length is 170 mm., the condylobasal length 180 mm.,
which may be the same dimension as the " basal length" of Nathusius.

In a series of nine skulls of Eskimo Dogs from Greenland, Baffin
Land, Labrador, Mackenzie, Alaska, eastern Siberia and Kamt-
schatka, collected for the most part many years ago, it is notable that
most are of about the same size as those of the Common Indian Dog.
One or two from eastern Siberia are the smallest and most slender.
All are heavy of bone, yet the sagittal crest does not show the strong
backward overhang seen in the Wolf's skull. The muzzle in most is
broad, yet this varies. The largest skull of all (U. S. N. M. 8,222)
collected by Dr. W. H. Dall at Nulato, Alaska, is nearly as long as a
small Wolf's, yet the teeth <^o not approach those of a Wolf in size.
This and other large skulls of Eskimo Dogs, probably are the result of
crossing with large dogs of European origin. Hearne (1796) speaks

ALLEN: DOGS OF THE AMERICAN ABORIGLVES. 445

of the large English dogs at the Fort on Hudson Bay; Ross (1S61)
notes the crossing of Eskiino Dogs Avith imported Pointers; and
Harmon (1820) records that by the early part of the last century,
large dogs imported from the English settlements of Newfoundland,
had already been introduced in the fur countries as far west as the
Rocky Mountains. It seems apparent that the large size of some
present-day Eskimo Dogs is therefore due to the influence of imported
stock, and that probably the aboriginal Plskimo Dog was not a much
larger animal than the Common Indian Dog. The thick coat, how-
ever, often adds much to its apparent size.

It seems to be somewhat characteristic of the Eskimo Dog that the
posterior narial opening (interpterygoid fossa) is broader and shallower,
less contracted at its rearmost portion, than in dogs of other breeds,
possibly correlated with their use in hauling and consec[uent need for
deeper breathing. In this respect, however, there is some variation;
\-et in certain larger skulls which are presumably of mongrel dogs, the
more narrowed and deepened fossa is ob\ious.

Thorndike (1911), in an interesting article on the Indian sfed-dogs
of North America, doubts if pure-blooded Eskimo or "Husky" Dogs
are today found in North America except possibly about the Copper-
mine River, Banks Land and Wollaston Land. "In general, the
Eskimo Dog differs from the Indian variety in beipg more wolfish and
in having less European strain. His tail is more bushy and he is
cleaner-legged. His ears are more erect and pointed, while his body
is larger in size" — this in comparison with the mongrel dogs of the
northern forest Indians of the present day.

Origin. — From its evident similarity of appearance to the Siberian
Sledge-Dog, it is generally accepted that the two are of similar origin.
The Sil)erian Dog seems indeed to differ in little except possibly its
slightly smaller size. Dogs of the same type are found across northern
Asia into Lapland, whence certain authors have concluded that the
Eskuno Dog was undoubtedly brought from the Old World by the
S^skimo themselves, who inust already have known how to use them
in harness. This \\v\y seems on the whole very probable. The
ultimate deri\ation of the Eskimo Dog and the so-called Spitz Dogs
in general, is howe^■er, still obscure. Some form of Wolf is commonly
looked to as the remote ancestor of the breed though direct proof is
not available. Holland (1908, p. 232) has even gone so far as to
suggest that certain well-preserved jaws discovered in a Pleistocene
cave-deposit at Frankstown, Pennsylvania, may from their i-esem-
blance to those of an Eskimo Dog, have come from a wolf-like ancestor

440 bulletin: museum of comparative zoology.

of this breed. The associated fauna, however, is of a more southern
character than would be expected as companions of this Arctic dog.

Of the larger dogs of the New World, the Eskimo Dog is the only
one that habitualh^ carries its tail curled forward over the hip. This
character, striking as it is, does not seem to have been particularly
studied from the standpoint of heritability, to see if it behaves as a
Mendelian character when contrasted with a drooping tail. Yet it is a
highly important trait, and is found not only among the dogs of similar
appearance in the north of Asia and Europe, but in other varieties,
possibly related, and of more southern habitat in those continents.
The so-called Chow Dog of China, a medium-sized red, or sometimes
black (Kreyenberg, 1910) dog, with erect earsand powerful shoulders has
the same sort of tail. A similar, though slightly smaller dog standing
50 cm. high at the shoulder is found among the Battaks of Sumatra
(Studer, 1901 , p. 31 ). The same curled tail is found in the Pomeranian
Dogs, that appear in the decoration of Greek vases (Keller, 1909) or as
figurines of Mycenian times. The fact that the curled tail carried
over the hip is so widely characteristic of certain breeds of Old World
dogs, where it seems to have been known from ancient times, implies
that it originated there and strengthens the view that the Eskimo
Dog came from Asia with the Eskimo. The contention that " the
canine of the American aborigine, or Amerind, was simply a tame
wolf, differing from its wild brother in the qualities that would nat-
urally follow breeding in the semi-domestication of the savage" and
that the dog "bred by the Indians in the forest regions, and the
Eskimos, was always derived from the Gray wolf" (Thorndike, 1911),
seems only remotely true. There is much evidence, though of a
somewhat uncertain character, that wild male Wolves will breed with
female Eskimo Dogs at proper seasons, and the northern Indians are
said to encourage such occasional crosses. Thorndike states that
tame wolves are sometimes seen in harness with the dogs in the North.
Nevertheless, under usual circumstances, those who have lived in
Arctic countries agree that wolves are highly unfriendly with the dogs,
and a single wolf is more than a match for several dogs. There seems
to be no evidence that Wolf cubs were habitually reared by either
Eskimo or Indian, which one would expect to be the custom if the
Eskimo Dog is merely a Wolf, tamed. Hearne (1796) mentions that
some Indians, on finding a Wolf's den, fondled the little cubs, and
painted their faces with vermilion, but returned them to the den and
made no attempt to rear them. He adds (p. 362) that " all the wolves
in Hudson's Bay are very shy of the human race, yet when sharp set,

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 447

they frequently follow the Indians for several days, hut always keep
at a distance. They are great enemies to the Indian dogs, and fre-
quently kill and eat those that are heavily loaded, and cannot keep up
with the main body."

A comparison of available skulls indicates that those of Eskimo
Dogs from eastern Labrador and western Greenland are constantly
smaller than those of eastern wolves, the teeth markedly smaller.
European investigators (Studer, 1901; Anutschin, 1881; Woldrich,
1882) have described skulls and other bones of large dogs from deposits
of the later Stone Age — Neolithic — one or two of which, the so-called
C. f. itiostranzcivi, C. f. Jadocjcnsis, seem to be large animals much like
Eskimo Dogs, and are considered as belonging to the same group.

Eiflfe (1909) records a crossing of the Australian Dingo with un
Eskimo Dog, in the Hamburg Zoological Gardens. The Dingo, a
female, was an unusually pale reddish brown animal; the dog, a black
East Siberian Sledge-Dog. The eight pups of this litter were more
reddish in color than their mother, with slightly bushy tails, somewhat
bowed upward. The old Dingo then paired with one of these reddish
dogs, and produced eight young, five very pale like herself, three
darker red. The ears of all the young were not at first erect, but
became so in the course of five months.

Notes. — The accounts of the early voyagers leave no doubt that
these large dogs were companions of the Greenlanders and American
Eskimo before the coming of Europeans. Their use l)y the natives
as sledge-animals makes them of prime importance in the Arctic
conditions under which they \\\e. Cranz and Egede, early Danish
missionaries to Greenland, mention the dog-teams, and the latter
author gives a crude figure. Scoresby in his Greenland Journal, (1823,
p. 203) relates finding at Jameson's Land in eastern Greenland, the
skull of a dog in a small grave, probably that of a child. The Eskimo
of this part of Greenland must have had very little contact with
Europeans up to that time. Cranz, in his History of Greenland,
alludes to this custom of the natives, who believe that by lading the
head of a dog beside the child's grave, the animal will shoAV the igno-
rant babe the way to the Land of Souls, for a dog can find its way
everywhere.

Among early accounts of the Eskimo Dogs, several of special inter-
est are given in Hakluyt's Voyages. In The second voyage of
Master Martin Frobisher, made to the West and Northwest regions,
in the yeere 1577 (Hakluyt's' Voyages. Everyman's Library ed.,
5, p. 137), it is related that a landing party at York Sound examined

448 bulletin: museum of comparative zoology.

the deserted tents of the Eskhnos, "not taking an;s' thing of theirs
except one dogge." The possessions of these people are described,
including " also dogges like unto woolves, but for the most part black,
with other trifles, more to be wondred at for their strangenesse,
then for any other commoditie needefull for our use." Again, " they
frank or keepe certaine dogs not much unlike Wolves, which they yoke
togither, as we do oxen & horses, to a sled or traile: and so carry their
necessaries over the yce and snow from place to place: as the captive,
whom we ha^'e, made perfect signes. And when those dogs are
not apt for the same use: or when with hunger they are constrained
for lacke of other \'ictuals, they eate them : so that they are as need-
full for them in respect of their bignesse, as our oxen are for us."
At Leicester's Island, in the present Frobisher Bay, a captive Eskimo
caught one of the Englishmen's dogs and showed how the natives
trained their animals. In the narrator's words, " Taking in his hand one
of those countrey bridles, he caught one of our dogges and hampred
him handsomely therein, as we doe our horses, and with a whip in his
hanfl, he taught the dogge to drawe in a sled as we doe horses in a
coach, setting himselfe thereupon like a guide: so that we might see
they use dogges for that purpose that we do our horses .... They drawe
with dogges in sleads upon the yce, and remoo\'e their tents there-
withall wherein the\' dwell in Sonuner." This seems to l)e the earliest
account of P^sknno Dogs in Arctic America by Englishmen. It is
interesting to find that the explorers carried a dog with them from
Europe, showing the possibility at an earl\- date, of contamination
of the breed with European dogs. John Davis, who sailed from
England in June, 1585, " for the discoverie of the Northwest passage,"
met with Eskimo Dogs in August, in Cumberland Sound. His
chronicler relates that here " we heard dogs houle on the shoare,
which we thought had bene volves, and therefore went on shoare to
kill them. When we came on land the dogges came presently to our
boat very gently, yet we thought they came to pray upon us, and
therefore we shot at them, and killed two: and about the necke of
one of them we found a leatherne coller, wjiereupon we thought them
to be tame dogs. There were twenty dogs like mastives with prickt
eares and long bush tailes" (Hakluyt's Voyages, Everyman's Library
ed., 5, p. 289).

At the present day, it is unusual to see typical Eskimo Dogs .south
of Hamilton Inlet on the Labrador east coast, though many mongrel
indi\'iduals are found about the settlements between there and New-
foundland. Three centuries ago, however, the natives of the latter

ALLEX: DOGS OF THE AMEHKAX ABORIGINE?!. 449

island had dogs wliich from their apparent resemblance to wolves,
may have been of the Eskimo breed. For Whitbourne, in his " Dis-
course and Discovery of Newfoundland" (London, 1622) writes that
the natives of Newfoundland " are a people that will seeke to revenge
any wrongs done unto them or their Woolves, as hath often appeared.
For they mark their Woolves in the eares with several markes, as is
used here in England on Sheepe and other beasts, which hath been
likewise well approved. For the Woolves in these parts are not so
violent and devouring as Woolves are in other Countries." The same
writer speaks with astonishment of his own mastiff's familiarity with
these tamed "Woolves" (Mercer, 1897), which it seems reasonable to
conclude were really Eskimo Dogs.

Of the Eskimo Dog in Greenland, BroMii (1868, 1875) considers the
breed to be practically the same as that of Davis Straits and Kamt-
schatka. In western or Danish Greenland he found it more or less
mixed with dogs of European descent and south of Holsteensborg not
used by the Eskimo, as the sea is not sufficiently frozen over in winter
for sledging. The same author adds that in 1861, Prof. Otto Torell
brought several dogs from Greenland for the use of his expedition in
Spitzbergen, where on account of the open water they were found
useless and set free. Within a few \'ears thev were said to have
increased in numbers.

Plains-Indiax Dog.

Characters. — Size mediuiii, slightly smaller than the Eskimo Dog;
ears large, erect; tail drooping or slightly upcurved; coat rather
rough, usually "ochreous tawny" or "whitish tawny," or sometimes
black and gray, mixed with white.

Distribution. — ^ Western North America from British Columbia
south perhaps to the Mexican Boundary and eastAvard through the
Great Plains Region.

Notes and Descriptions. — It is apparently to this dog that most of
Lord's description (1866, 2, p. 222) applies in his Naturalist in Van-
couver Island and British Columbia. So impressed was he by the
general similarity of these dogs to coyotes, that he believed the one
derived from the other, and makes one general description do for
both, with the addition that in the dog the hair "becomes shorter,
softer, and more imiform in coloration, although the tail retains its
bushy appearance." The general color is an "ochreous grey," the
liairs tipped with black, those of the neck tricolored, ha\ing theii

450 bulletin: museum of comparative zoology.

"lower two-thirds reddish brown; then a ring of white, and a black
tip." This pattern gives "a most curious speckled look" to the
bristling neck of an enraged dog. Coues (1873) was equally impressed
by the general resemblance of these dogs of the Plains Indians to
coyotes and considered the two animals essentially the same in struc-
tural points, though he thought if" unnecessary to compare the skulls."
Indeed, he accepted it as unquestionable that in every Indian com-
munity mongrel dogs are found, shading into coyotes in every degree.
Such crosses he says, are obtained by picketing female dogs over night
at proper times, thus allowing them to cross with coyotes. Morton
(1851) quoting a letter from Dr. Cooper, Fort Duncan, Texas, speaks
of ever}- ranch having a dog resembling a coyote, "and a bitch to
which no dog had had access, produced whelps, evidently a cross with
the Coyote^ Wortman, also (in Cope and Wortman, 1884, p. 8, foot-
note) after extended travel in the western United States corroborates
Coues — but from hearsay evidence, however. He found among the
Umatillas, Bannocks, Shoshones, Crows, Arrapahoes, and Sioux,
mongrel dogs, " which to one familiar with the color, physiognomy
and habits of the coyote, have every appearance of blood relationship,"
if they are not " in many cases, this animal itself in a state of semi-
domestication." All such evidence, however, is unsatisfactory, and
rests on general resemblances in fonn, color, and characteristics that
may be common to both animals. A comparison of skulls and teeth
would perhaps reveal more significant tokens of the true relationship,
but hitherto nothing has been published as to the cranial characters
of such animals. Yet, in his much-quoted paper on the origin of the
American varieties of the dog, Packard (1885) appears to have been
influenced by Coues's belief, and agrees with him in considering these
dogs as merely tamed coyotes. In a journey through provincial
Mexico he was struck by the general resemblance of the native dogs
to these animals, and again, in 1877, on the upper Missouri took
special note of the dogs of the Crow Indians, describing them as of
wolf-like appearance, of the size and color of a coyote — a whitish
tawny — but less hairy and with less bushy tails. Lord (1866, 2,
p. 221) found a number of dogs with a little tribe of Indians at Sweltza,
a small lake west of the Cascades, near which the boundary of British
Colmnbia passes, " that were hardly in any degree altered from the
cayote" in exterior appearance. He speaks of their burrowing
deeply into the ground to bring forth their young, but this trait is
found in dogs as well as in coyotes. From these accounts it is clear
that the general appearance and coloration of this dog are strikingly

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 451

like those of one of the coyotes. Hamilton Smith (1840, p. 156)
refers to the same dog as the " Techichi of Mexico, or the Carrier-dog
of the Indians," and gives a figure (PI. 4) of the only example he had
seen, a tawny dog of normal proportions and with cropped ears. He
confuses it however, with Richardson's "Carrier-Indian" or Short-
legged Dog and further complicates his account by supposing it the
same as the Mexican Techichi.

James Teit (1909) writing of the Thompson Indians of the upper
Fraser River, British Columbia, also remarks on the general resem-
blance of their dogs to coyotes, but adds that through intercrossing
with dogs imported b;\' the whites, the breed has become totally
extinct. They were good hunters, though poor watch-dogs, and the
l)est ones for deer hunting were highly prized. Such dogs generally
ran the deer to water, often bringing it to bay in some creek, and keep-
ing it there till the Indian came up and dispatched it.

It is regrettable that more thorough comparison of the teeth of these
dogs could not be made to test any supposed resemblance or relation-
ship to coyotes. As Gidley (1913) has pointed out, the fourth lower
premolar of the latter has normally two secondary cusps and a cingu-
lum, that of the dog normally but one secondary cusp, a ready means
of distinction in addition to other relative characters. It should be
added that in numerous fragments I have examined from the south-
west, there is no evidence of coyote influence.

Referable to this same breed are perhaps the larger dogs mentioned
by Suckley (Suckley and Gibbs, 1860, p. 112) as kept by the Indians
" about the Dalles of the Columbia," Oregon. These he describes as
about the size of a foxhound, but much more slender, in color yellow
or brindled.

A similar type of dog seems to have been kept by the Indians of
California. At all events, a series of skulls from mounds on the south-
ern coastal islands are hardly to be distinguished from New Mexican
skulls. A skull found in association with that of an Indian, washed
out after a freshet, from a bank at the junction of the Tuolumne and
San Joaquin Rivers, California, is of the same medium-sized type,
rather heavy of bone, slender of muzzle, and with feel)le sagittal crest,
mainly on the occiput.

Skeletal Measurements. — A cranium discovered in the course of
excavations by Dr. A. V. Kidder at Pecos, New Mexico, may be
attributed to this dog. It is nearly identical in size and proportions
with several of the skulls from southern California from mounds on
the island of San Nicolas, kindly loaned me by the Archaeological

452 bulletin: museuim of comparative zoology.

Department of the Unhersity of ( 'alifornia. Tliese last are in an
excellent state of preservation, of medium size, yet of massive bone,'
with roughened brain-case, and sagittal crest developed mainly on
the interparietal region. The teeth are rather small, the first upper
premolar lacking in some cases.

The following table gives the cranial measurements of several of
these skulls. The first two, from Pecos, N. Mex., differ in that the
one, a rostrum only, is considerably larger than the other, or any of
the Californian skulls. Of the latter, there are several from mounds
on San Nicolas Island, which represent a dog apparently identical
with that of New IVIexico. The last tAvo columns giA-e dimensions of
two old dogs with much worn teeth; in the larger, indeed, the upper
molars have been lost and their alveoli partially filled, while the remain-
ing teeth are mere stumps. The smaller of these two skulls, while
not very different in the measurements of the tooth-row, has a shorter,
smaller cranium. It is very likely a mongrel betAveen this larger dog
and one of the short-nosed dogs (' Pachycyon'), a relationship further
indicated by its slightly more upturned snout. It is further peculiar in
lacking the first upper premolars on both sides, Avhile in the lower jaw
there are on both sides four molars, the second and third each Avith
two roots and the fourth single-rooted like the usual third molar.
Four molars in the lower jaw is not an unknown featin-e in the dog.
Nehring (1882) found twentA' dog skulls out of 650 in Avhich there was
an extra molar either in both upper or both lower tooth-rows, or in
only one tooth-row.

Lucas (1897) has giA^n a brief account of the cranium of a large
dog, evidently domesticated, found in an ancient Pueblo Indian graA-e
at Chaves Pass, Arizona, in 1896. Another of similar proportions
was discoA'ered at San Marcos, Texas, associated with Hints, a human
skeleton, and other bones. The former skull he regards as of a " broad-
faced type," and describes it as "precisely similar in size and pro-
portions to the cranium of an Eskimo dog from Cumberland Sound."
He supposes these to be carrier-dogs, and recalls Clavigero's mention
of them as "a quadruped of the country of Cibola [New Mexico],
similar in form to a mastiff, which the Indians employ to carry bur-
dens." I have not been able to examine these skulls, but they maA' be
the same as the larger of the two Ncav ]VIexico skulls here listed.

ALLEX: DOG« OF THE AMERICAN ABORIGINES.

453

•

-8

Measureinenls of Ihc Skulls

_; ^

a ■

Oceipitonasal length (exclud-

ing incisors)

— •

173

170

164

172

178

159

Basal length

—

153

151

146

153

156

143

Palatal length

Median length of nasals

—

Alveolus of i^ to anterior edge

of orbit

Alveolus of i^ to 7n-

86.5

89.

91.5

" " canine to ni-

" /ji to m'-

59.5

—

" 7J2 to nf

" jr' to m'

—

" p' to m^

Alveoli of i?i^ and m-

16.5

Length of carnassial (p*)

17.5

—

Width of occipital condyles

—

34.

" " palate at »t'

(>4

" across supraorbital

processes

—

Zygomatic width

—

106

112

111

Lower jaw, alveolus of ii to itig

—

"canine

to >nz

—

Lower jaw, alveolus of pi to vi:i

—

— •

"?>2t0 7H3

—

"p3tO/«3

—

■■ —

—

"?rt,tom3

—

Length of nii (carnassial)

—

■ —

—

Uses. — These dogs of medium size, were chiefly used by the Indians
in transportation, secondarily in hunting. In the plains country
from Saskatchewan to the Mexican Boundary, the travois was in
general use. This consisted of two light poles, the smaller ends
fastened together and resting on the dog's shoulders, the heavier ends

454 bulletin: museum of comparative zoology.

kept apart by a crosspiece and trailing behind. A leather collar served
to keep this frame in place for dragging the goods piled upon it. In
this way entire villages moved, the dogs dragging the household
effects. The contrivance seems not to have been used west of the
Rocky Mountains. Perhaps the earliest mention of the use of these
dogs as pack-animals is found in Coronado's account of his journey
in 1540 to 1542, from the City of Mexico to the Texas plains (see
translation l)y Winship, G. P., 1904). When some ten days' march
from the present Rio Pecos, Texas, Coronado and his followers came to
Haxa, where the natives were found to have "packs of dogs." In
moving camp, these Indians started off "with a lot of dogs which
dragged their possessions." "They travel like the Arabs, with their
tents and troops of dogs loaded with poles and having Moorish pack
saddles with girths. When the load gets disarranged, the dogs howl,
calling some one to fix them right." A letter from one of Coronado's
men further describes the dogs. "These people," he writes, "have
dogs like those in this country [Spain], except that they are somewhat
larger, and they load these dogs like beasts of burden, and make saddles
for them like our pack saddles, and they fasten them with their leather
thongs, and these make their backs sore on the withers like pack
animals .... When they move — for these Indians are not settled in
one place, since they travel wherever the cows [i. e., Bison] move, to
support themselves, these dogs carry their houses, and they have
the sticks of their houses dragging along tied on to the pack saddles,
besides the load which they carry on top, and the load may be, accord-
ing to the dog, from 35 to 50 pounds." Evidently these were the
carrier-dogs of the Plains Indians, and the method of packing with the
tent poles used as travois seems to be here first described.

As pack-animals, for moving camp in their pursuit of the Bison,
these dogs were of great service to the Indians of- the plains country,
and every village was provided with troops of them.

As an article of food, the dog seems to have been somewhat analo-
gous to the fatted calf. George Catlin (1841, 1, p. 14) writing of the
Upper Missouri Indians, says: "We are invited by the savages to
feasts of dog^s meat, as the most honourable food that can be presented
to a stranger."

allen: dogs of the american aborigines. 455

Sioux Dog.

Characfcr.s. — A large wolf-like dog, probably elosely related to the
Plains-Indian Dog but larger and gray rather than tawii}- in color.

Distrihvtion. — Probably the north-central plains area, from the
Missouri north perhaps to Saskatchewan.

Notes. — No doubt the carrier-dogs differed slightly among the
various tribes of Plains Indians co\'ering the wide stretch of country
from Northern Mexico to Saskatchewan, so that local breeds of the
general type could be distinguished did we have opportunity to com-
pare them. Morton (1851), who tried to obtain information from
frontier officers in the earlier half of the last century, cjuotes a letter
from H. H. Sibley, a correspondent in ^linnesota, who avers that
" the Indian Dog differs much in size and appearance among different
tribes" but that they all have small, sharp, erect ears. He particu-
larly recalls that "among the Sioux, it is large and gray, resembling
the Buffalo Wolf." Packard (1885) has mentioned "whitish tawny"
Indian dogs seen in 1877, among the Crows of the upper Missouri.
Lewis and Clark, on their famous journey, came upon a scaffold
burial of an Indian squaw, near which lay two dog-sleds and the
carcase of a large dead dog, between Mandan and the Yellowstone.
These large gray dogs of the Sioux may ha^e been a distinct breed
from the tawny dog, of the size of a Coyote, and possibly the same as
certain large dogs seen by Hind (1859) among the Crees of the Sand
Hills. Sir John Franklin (in his Journey to the shores of the Polar
Sea, 1829, 1, p. 176) briefly mentions the large dogs of the Crees in the
Saskatchewan countrv. He adds that in the month of March, the
female wolves "frequently entice the domestic dog from the forts,
although at other seasons a strong antipathy seemed to subsist between
them."

Hamilton Smith (1840) quotes an interesting letter from Prince
Maximilian of Wied, likening- the North American plains dog to a
wolf, "excepting that the tail is more curved, and the color either
"absolutely grey like wolves" or white, black, and black and white
spotted. The latter coloring, however, may apply to some other
breeds than that under consideration.

Figures probably representing this dog, are shown in some of the
plates of Catlin's Indians (1841, colored edition, 2) small to be sure,
but showing the gray coloring, large erect ears, and scimitar-shaped
tail carried out behind. His Plate 103 in 2 is a spirited drawing
illustrating a dog-fight in which all the dogs of the party, though
burdened with their loads "en iravois," are rushing to participate.

456 bulletin: museum of compakative zoology.

Long-haired Pueblo Dog.

Characters. — A medium-sized dog of slender muzzle, erect ears,
and normal bushy tail. Hair long and dense, pale yellowish, clouded
with dark brown on ears and croAvn, whitish beneath on throat, belly,
and feet. Feet well-haired. Probably this is to be looked upon as a
local breed of the Plains-Indian Dog, from which it apparently differs
only in its longer coat.

Disiribuiion. — Known only from the Marsh Pass region of Arizona,
but in former times probabl}' common to the Pueblo tribes of Arizona
and New Mexico.

General Account. — One of the remarkable discoveries of ^Messrs.
Guernsey and Kidder, Avhile exploring for the Peabody Museum,
was an excellently preserved specimen of a medium-sized dog associ-
ated with a human burial. In the arid climate of Arizona, the
dog had merely dried, so that the entire animal even to the thick hair
was nearly intact. It is co\'ered with a dense coat of long woolly
hair, of a pale yellowish color, clouded on the l)ack and head with
brownish. On the sides of the body, the length of the hair is about
100 mm.; on the toes 30 mm. The culture period to which this
specimen belongs, is believed by Mr. Guernsey to antedate that of the
Cliff Dwellers, and hence must be at least several centuries old.

It seems probable that it was to this long-haired flog that Mendoza,
a companion of Coronado, refers in a letter of 17 April, l.")40, to the
King of Spain, describing the pueblo of Cibola, then a famous Indian
site, near the present town of Zuni, New Mexico. This letter is trans-
lated by Winship (1904, p. 153) from the Spanish of Pacheco y Car-
denas, (Documentos de Indias, 2, p. 3.50), and contains the following
passage:— "In their houses they keep some hairy aniinals, like the
large Spanish hounds, which they shear, and they make long colored
wigs from the hair, like this one which I send to Your Lordship, which
they wear, and they also put this same stuff into the cloth which they
make." These "hairy animals, like the large Spanish hounds,"
seem probably, in the light of Mr. Guernsey's discovery, to have been
the same as the dog found at Marsh Pass. It is recalled here that
breeds of long-haired dogs were kept for shearing not only by the
Indians of Puget Sound, but by the Chonos of the Taitao Archipelago,
Chile, and their hair woven into blankets (see p. 475). There was
fonnerly a breed of long-haired white or brown dogs among the
aboriginal inhabitants of New Zealand, the prodiict of which was
similarly u.sed (Colenso, 1878).

ALLEX: DOGS OF THE AMERICAN ABORIGINES. 457

External Measurements. — It is not possible to remove the skull and
limb-bones without injuring the mummy for exhibition purposes.
A few dimensions, however, follow: —

Length from nose to root of tail, following backbone — about 700 nun.

Length of tail, (bnjken at tip) slightly over 200

Hind foot ' 141

Femur (approximately) 14o

Tibia (approximateh-) 143

Upper jaw, front of canine to back of pm* 55.5

l^'pper carnassial ipm*) 18

Length of skull from occiput to tip of nose (.appro.ximately) . . 195

Width outside upper canines 31

" " carnassials 54

Zygomatic width — about 95

Lower jaw, front of canine to back of mi 68 . 5

« u u n 11 ii « a AQ

" '■ jitHi to pnii 35

Length of lower carnassial 21

Larger or ("ommox Indian Dog.
Plates 7, 8. .

1817. Cants fdniiliiiris <uncricanus canadensis Walther, Hund, p. 43.

1829. Canis faniiliaris YAr. c. canadensis Richardson, Fauna Boreali-Amer., 1,

p. 80 (not Canis lupus canadensis Blainville 1841, which is Canis lycaon

Schreber) .
1834-6. Canis canadensis Reichenbach, Regn. anim., i)t. 1, p. 46, fig. 564.
Canis familiaris urtholu^ cniiadcnsis Reichenbach, Naturg. raubth..

p. 146, fig. 564.
1867. Canis domesticus borealis luparius Fitzinger, Sitzb. K. akatl. wiss. Wien,

56, pt. 1, p. 409 (not C. ./'. nrt-hotus Iupariu>i Reichenbach, Regne anim.,

pt. 1, p. 13, fig. 131; not Cants donieslicu-i luparius Fitzinger, Sitzb.

K. akad. wiss. Wien, 1866, 54, pt. 1, p. 406; 1867, 56, pt. 1, p. 396.
1881. Canis lutrans domesticus Langdon, Journ. Cine. soc. nat. hist., 3,

p. 299 (not Cam's ftnniliaris domesticus Linne, 1766).

Characters. — This was probably closely related to the Plains-Indian
Dog, but seems to have been usually solid black or black and white
in patches instead of resembling the Coyote in color. The skull has,
when adult, a knife-like sagittal crest, a high forehead, and is rather
slender. Limbs much longer than in the Short-legged Indian Dog

45S bulletin: museum of comparative zoology.

yet slightly inferior to those of a Greyhound. The first lower pre-
molar was frequently wanting.

Distrihidion. — Dogs of this general tjy-pe, agreeing fairly well in size
and proportions were found among the forest Indians from Alaska
southward to Florida and the Greater Antilles, and westward to the
edge of the plains in the east central States. The more northern dogs
seem to average a little larger than those from the south, but in the
absence of more exact knowledge seem best referred to this type. No
doubt in the far Northwest there was more or less mixture with the
Eskimo Dog. Probably too, local strains of this general type of dog
could be distinguished, did we know their external characteristics,
but the skulls and teeth seem remarkably similar over a wide area.

Skeletal remains. — Cope (1893) was the first to describe the jaw of
this dog from a specimen collected by Moore from a shell-mound on
St. John's River, Florida. He was struck by the fact that the first
lower premolar was missing and appeared not to have developed.
The strong development of the entoconid of the carnassial, he also
noticed. Moore, in the course of various explorations in Florida and
Georgia discovered many remains of dogs, apparently of this t^-pe.
In a large mound on Ossabaw Island, Georgia, he (1897) found several
interments of human and dog-skeletons, the latter always buried sepa-
rately and entire, showing that the dogs had not been used as food.
Other dog-skeletons of a similar sort were found by Moore (1899) in
aboriginal mounds on the South Carolina coast. Several of the
skulls collected by him are in the Peabody Museum, where I have
had the privilege of studying them. Putnam (1896) considered them
the same as those of the larger Madisonville dogs. More recently
the M. C. Z. has received from Prof. Carlos de la Torre, two frag-
mentary skulls of dogs associated with pre-Columbian burials in Cuba.
These skulls seem to be essentially similar as far as can be judged.
Miller (1916) has reported a lower jaw of a dog from an Indian site
in Cuba.

Three crania in excellent condition, from the INIadisonville, Ohio,
site agree in their somewhat slender proportions, with narrow palate
and rostrum. A strong but thin bony crest is developed along the
midline of the brain-case, and there is a noticeable inflation of the
region just back of the supraorbital processes. The first premolar
is absent in both eraniimi and jaw of one specimen. Two crania from
a shell-heap at La Moine, Maine, similarly lack the first premolar.
One of these latter is a much larger skull than any of those from
MadisonA-ille, which may indicate some variation in the local breeds.

ALLEN: DOGS OF THE AMERICAN ABORIGINES.

459

yet the general type seems to be the same. Hardly distinguishable
from the Maine specimens in any way is a skull from Peel River,
Yukon, (U. S. N. M. 6,219) collected about 1860 by Kennicott and
representing probably the common Indian Dog of that region.

Cruiiial Measuremeuls

Alveolus of i' to occipital condyle .

Median length of nasals

Alveolus of i^ to median edge of

palate

Alveolus of i^ to anterior edge of

orbit

Alveolus of i^ to m-

" " canine to ni-

" " pi to w2

" pUo m2

Alveoli m^ and ni^

Lengt h oi p^

Width of occipital condyles

" " palate at ni^

" across supraorbital

processes

Zygomatic width

.2§

^ .
o a,

170
56

72.5

18.2

50
102

■a
a

172
62

77.5

62.5

37.5

o
•5

O &H

163

87
72

20.8

18.6

47
104

177

81
96
79

62.5

20.5

66.5

49
101

c oo'

163
57

17.5

46
92

^ CO

c .

•: §

cs .

169
57

18.5

57
104

74
64
55
16.3

38
62

192 ±

20.8

168

70 ±

56
19.8
19.7
37

Of seven lower jaws from Maine shell-heaps, all but one lack the
first premolar, and the same tooth is lacking in a ramus from Madison-
ville. It seems to be missing in the greater portion of lower jaws of
this dog. The following measurements show the lengths of different
parts of the tooth-row taken at the alveolar borders, because the
teeth themselves are frequently lost.

460

bulletin: museum of comparative zoology.

—

■■■

Tooth-row Measurements

C t

4. 5.

- ^

C >^

C >i

c >>

C "^

■« ^

■5 s

ro "Z

'Z —

■2 ~

•- cc

^-J;

^ tz

S X

i ;:

> i.

5 ^

Alveoli, ii to Ws

—

100

—

105

" c to tris

—

" p2 to ms

72.5

71.5

" pa to ms

—

" Pi to ms

—

" 7«i to Wis

33.5

Length of tooth, mi

22.5

22.3

21.5

Skekial Meafiiironcnts. — The first of the Calf Island jaws al)Ove, is
accompanied by parts of the skeleton of the same animal. The limb-
bones of this skeleton and those of several dogs from Madisonville,
Ohio, measure:

•

_c;

.= ^ '

c -c

- -c

" a

'-C £

.£ '"

_ —

^ —

S 0

C ^

C ?

—

Huineru.s

168

163

162

Radius

164

—

164

163

• — •

—

Femur

170 ±

—

— ■

—

173

—

Tibia

172

—

177

160

156

Notrs and DcbTriptions. — On account of the finding of cranial
fragments that appear to represent this animal, in aboriginal l)urials
in Cuba, it is assumed that this is the dog mentioned by the first
discoverers under Columbus. Oviedo (1535) writing of the aboriginal
dogs in Haiti shortly after the discovery, declared that they were no
longer to be found there in 1535, as all had been killed for food during
a time of famine. These dogs he described as of all the colors found
among the dogs of Spain, some uniformly colored, others marked with
blackish and white, or reddish brown. The coat of some was woolly,
of others silk\' or satiny, but most of those in Haiti were between silky
and satiny, yet rougher than the Spanish dogs; with ears pointed and

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 461

erect like those of wolves. None of these dogs barked. Oviedo
adds that similar dogs were plentiful in many parts of the continent,
as in Mexico, Santa Marta, and Nicaragua. He had eaten their
flesh and considered it excellent, resembling lamb. In Nicaragua
and Mexico the Indians bred numbers of them and at their great
festivals dog-meat was considered the best dish of all. The natives
of Haiti hunted some species of Hutia with these dogs.

Very little seems to have been written descripti\'e of this breed.
In his essay on the origin of dogs, Hunter (1787) mentions that a Mr.
Cameron, who had lived among the Cherokee Indians, informed him
that the dog found in their country was "very similar to the wolf."
Cameron thought it remarkable there were not sundry breeds of dogs
among these Indians, as in Europe. William Bartram (1792, p. 220),
during his travels in Florida, made special note of a " single black dog,
which seemed to differ in no respect from the wolf of Florida, except
his being able to bark as the common dog." It belonged to an Indian,
who had trained it to tend a troop of semiwild horses, " keeping them
in a separate company where they range; and when he is hungry or
wants to see his master, in the evening he returns to town, but never
stays at home at night." Barton (1805) appears to have made more
particular inquiry of Bartram concerning these Indian Dogs of
Florida, and describes them as " very similar to the Canis Lycaon, or
black wolf," yet they are not always black "but of different colours,
commonlv of a bav colour, and about one third less than the wild
black wolf. It carries its ears almost erect, and has the same wild
and sly look that the wolf has." Barton adds that the dogs of the
Cherokees were already (1805) much intermixed with the European
dogs.

Peter Kalm informed John Bartram that the dogs of the Canadian
Indians (?Montreal) were like those in Sweden with erect ears, and
Bartram himself (in a letter to George Edwards, 1757) recalled as a
boy seeing the Indian Dogs, with erect ears, accompanying their
masters on occasional visits to his father's house in Pennsylvania.
Barton (1805), who seems to have made diligent inciuiry about these
dogs, further describes their aspect as "much more that of the wolf
than of the common domesticated dogs. His body, in general, is
more slender than that of our dogs. He is remarkably small behind.
His ears do not hang like those of our dogs, but stand erect, and are
large and sharp-pointed. He has a long, small snout, and very sharp
nose." This breed, he says, was still preserved in the greatest purity
among the Six Nations, from whom the Delawares acknowledge that
thev received it.

462 bulletin: museum of comparative zoology.

Judging from the numerous shell-heap remains of what seems to be
this same dog, it was formerly common among the New England
Indians. In Hakluyt's Voyages (Ever^y-man's Library ed., 6, p. 95)
is an account of The voyage of the ship called the Marigold of Mr.
Hill of Redrife unto Cape Briton and beyond to the latitude of 44
degrees and an half, 1593. The narrator tells of meeting with a
party of " Savages" at Cape Breton in July, who upon the accidental
discharge of a musket, came " running right up o\er the bushes with
great agilitie and swiftnesse. . .with white staves in their handes
like halfe pikes, and their dogges of colour blacke not so bigge as a
grej'hounde followefl them at their heeles; but wee retired unto our
boate."

It is probably to this breed of dog that Charle^'oix refers in his
Journal of a voyage to North America (London, 2 vols, 17(U, transl.).
"The Indians," he writes, "always carry a great number of dogs with
them in their huntings; these are the only domestick animals they
breed, and that too only for hunting; they appear to be all of one
species, with upright ears, and a long snout like that of a wolf "
(1, p. 187).

This is the "major" type of Indian dog reported by Loomis and
Young (1912) from Maine shell-heaps, where rather large-sized speci-
mens have been discovered. Dog-remains have been found also in
Connecticut (MacCurdy, 1914) and Block Island, R. I. (Eaton, 1898).

An Indian Dog-skvill (Plate 7) collected by Kennicott on the Peel
River, about 1860 (U. S. N. M. 6,219) is hardly different, except for
its very slightly greater size, and seems best referred to the same sort
of dog, though possibly a distinguishable breed. Richardson (1829)
named this dog Canis familiaris var. canadensis, and says it is the
kind "most generally cultivated by the native tribes of Canada and
the Fur countries." He describes it as intermediate in size and form
l)etween the Eskimo and the Hare-Indian Dog. Its fur is black and
gray, mixed with white; some are all black. Apparently identical
with the skull from Peel River is another collected by Dr. W. H. Dall,
from a prehistoric Aleut village site in Unalaska. Dr. Dall notes that
this is the only dog-skull which had been found in the undeniably
prehistoric kitchen-middens of the Aleutian Islands. It still retains
the upper carnassial, which measures 20.5 mm. in length. The
occipital condyles are 38 mm. across. The first upper premolar was
apparently lacking.

Probably it was a dog of this breed that Audubon figured as. the
Hare-Indian Dog, from a living one in the gardens of the Zoological

ALLEN:. DOGS OF THE AMERICAN ABORIGINES. 4(i^>

Society of London. Bernard R. Ross (1S61) seems to have confusecf
the two as well; for a skull collected by him at Fort Simpson and sent
to the U. S. N. M. as ''Cam's laqopus'" is even larger than the one
from Peel River and almost undoubtedly a cross with an Eskimo
Dog. Both skulls lack the first lower premolar.

In the North the Common Indian Dog is largely used among the
forest Indians as a l)east of burden.

Samuel Hearne, on his famous journey to Peel Ri^'er, 1769-72,
observed that the Indians' " kettles, and some other lumber, are
always carried by dogs, which are trained to that ser\ice, and are
very docile and tractable. * *. * These dogs are equally willing to haul
in a sledge, but as few of the men will be at the trouble of making
sledges for them, the poor women are obliged to content themselves
with lessening the bulk of their load, more than the weight, by making
the dogs carr^• these articles onh-, which are alwa\s lashed on their
backs, much after the same manner as packs are, or used formerly to
be, on pack-horses."

Klam.\th-Indian Dog.

Characters. — A medium-sized dog, with a .short, bushy tail.

Distribution. — So far as known, this peculiar breed was found only
among the Indians in the Klamath River region of Oregon.

Remarks. — The only mention of this dog that I have found is the
following by Gibbs (Suckley and Gibbs, 18(30, p. 112):

"On the Klamath is a dog of good size, with a siiort tail. This is
not more than six or seven inches long, and is bushy, or rather broad,
it being as wide as a man's hand. I was assured they were not cut,
and I never noticed longer tails on the pups. They have the usual
erect ears and sharp muzzle of Indian dogs, but are (what is unusual
with Indian dogs) often brindled gray.'"

Presumably the shortened tail arose as an independent \ariation
among dogs of the Plains-Indian Dog type and was preserved among
these dogs through selective breeding. Similar short-tailed breeds
are well known among European dogs, as in the English Sheep-dog,
and certain varieties of Bull-terriers. ^lacFarlane (1905, p. 096)
gives an accovmt of a very much prized Eskimo Dog he owned in the
^Mackenzie District, that was born tailless and undersized, but pro\ed
an excellent sled-dog.

41)4 bttlletin: museum of comparative zoology.

Short-legged Indian Dog.
Plate 5, %. ].

1829. Canisfamiliaris var. d. novae caledoniac Richardson, Fauna Boreali-

Amer., 1, p. 82.
(?) 1912. Cards familiaris, minor Indian dog, Loomis and Young, Amer.

joiirn. sci., ser. 4, 34, p. 26, fig. 4, D.

Characters. — Ears erect, liead large in proportion, and body long;
the legs relatively short but not distorted as in our Turnspits. Fur of
the body short and sleek, that of the tail longer. This is possibly a
derivative of the Common or Larger Indian Dog.

Disiribution. — It is hardly possible to trace the former distribution
of this type of dog. It was found by Ptichardson in southern British
Columbia, and a dog apparently similar is known from Quebec, and
perhaps formerly in New England and New York. Probably it was
found among canoe-using or forest-li\'ing tribes in the North, hence
was infrequent or absent in plains country.

Notes and Descriptions. — Apparently Ptichardson (1829) was the
first to take special note of this breed. He found it among the Attnah
or Carrier Indians of "New Caledonia," (now British Columbia) and
it seems to have been bred as well by neighboring tribes as far south at
least as northern California. For Gibbs (Suckley and Gibbs, 18f)0,
p. 112) makes particular mention of seeing "one peculiar looking dog
on Eel River, in the interior of northern California, among very wild
Indians. It had short legs and long body, like a turnspit." Suckley
in the same work, briefly says that " the Indian dogs about the Dalles
of the Columbia [Oregon] are so xaried in appearance that no special
description can be given. We might, however, make two types. The
large * * * and the small, resembling the ' turnspit kind ' of which Mr.
Gibbs speaks. The latter are generally white, or spotted liver and
white, or black and white. This kind is kept more as a playmate for
the children and a pet for the women."

It is significant that Suckley mentions the "varied" appearance of
the Oregon dogs, so that it was possible to refer them in general to
but two types. This may have been a result in part of the inter-
breeding of the larger and the smaller types, and in part perhaps of a
mixture as Suckley suggests with European breeds already intro-
duced.

ALLEN; DOGS OF THE AMERICAN ABORIGINES. 4()0

Although generally associated with the Indians of British Columbia
and neighboring parts of the northwestern United States, it seems
likely that this or a similar breed may have been much more widely
distributed over northern North America, as far east and south as
Quebec, New P^ngland, and New York, if not farther. An excellent
photograph given me by Mr. W. B. Cabot (Plate 5, fig. 1) was ob-
tained a few years since among the Bersimis Indians, Quebec, and
seems to represent a dog of the same general type. The large head,
erect ears (somewhat laid back in the photograph), long, heavy Iwdy,
short, straight legs, up-turned tail, agree well with other descriptions.
This particular individual has the spiritless air of an old dog.

That this breed of dog was found at least as far south as the south-
ern coast of New England, may possibly be inferred from the account
by Livermore (1877, p. 58) of the dogs of the Block Island Indians,
of Rhode Island. This isolated colony of Indians numV)ered some ;^00
individuals up to the year 1700, but by 1774 was reduced to only ")1.
In 1876, there was known to be but a single one living on the island.
According to the author just mentioned, "the 'dogs' of Block Island
belonging to the Manisseans before the English came have their
descendants here still, it is belie\-ed. They are not numerous, l)ut
peculiar, differing materially from all the species which we ha\-e
noticed on the mainland, both in figure and disposition. They are
below a medium-size, with short legs but powerful, broad breasts,
heavy quarters, massive head unlike the bulldog, the terrier, the hound,
the mastiff, but resembling mostly the last; with a fierce disposition
that in some makes but little distinction between friend and foe."
The description here given, unsatisfactory though it be, implies a dog
much like that shown in fig. 1, Plate 5.

Skeletal Revmins. — I am unaware of the existence in any museum,
of bones that may be definitely associated with the Short-legged
Indian Dog. But, as pointed out by Loomis and Young (1912),
there are in the prehistoric shell-heaps of the New England coast
remains of a larger and a smaller Indian Dog, the latter of which on
the strength of the evidence just given as to the fonner presence of the
short-legged breed in eastern Canada and New England, may tenta-
tively be referred to this animal. The authors nientioned ha\-e cliar-
acterized the lower teeth of this smaller dog on the basis of jaws from
the Maine shell-heaps and through the kindness of Professor Loomis
I ha\e had opportunity to study the specimens.

The mandibles are all more or less broken, but include several in
fairly good condition. Thev differ from those of the Larger or Com-

466

bulletin: museum of comparative zoology.

men Indiun Dog in the .smaller size of the individual teeth as well as
in the shorter tooth-row. Yet the contrast is not al\va\'s very strik-
ing and no doubt there was more or less intercrossing of the two types.
The teeth of the smaller dog are usually more close-set than those of
the larger, and on comparison, the carnassial tooth is seen to be de-
cidedly smaller, its metaconid sometimes quite obsolete, and with a
distinct tendency for the outer of the two cusps of the heel (hypo-
conid) to become enlarged and trenchant. As in the Common Indian
Dog, and in American aboriginal dogs generally, it is common if not
usual, for the first lower premolar to be lacking, and the same is
frequently true of the first upper premolar. Such an anomaly is
occasional in all domestic dogs. Indeed, Bourguignat (1875) founded
his genus Lycorus on such a fossil canid jaw — probably of a wolf —
from a cavern-deposit in France. In his specimen the first premolar
was lacking in each ramus.

Mea-surfiiicnts of the lower

.juws and frafjments of

upper maxillae

Me.

FlafCKls.

S 1

98.5

1209

C 1

C'2

1S:5

Greatest length of lower
carnassial

Number of lower pre-
molars

Alveolar length p-, to m-i
" ]h to Pi

Alveoli, upper p"-m''-

Greatest length of y>'
(tooth)

39 . 5

17.3

19.8

65.5
33

20.3

3
31.5

3
68 J
34

3
65
32.5

20 . 6

3
64
32

21.3

4
66
33

Loomis and Young (11)12) mention similar small jaws from Indian
sites in Arkansas.

Of limb-bones referable to the Short-legged Dog it is particularly
desirable to obtain specimens for comparison with the other breeds.
Among limb-])ones in the Amherst collection from Maine are several
longer and shorter. The latter in the lack of evidence to the con-
trary, may be regarded as ha\'ing come from the present type. Of
two humeri, one is nearly perfect and appears to be that of an adult
animal, with its epiphyses throughly fused to the shaft. Its ole-

ALLEN: DOGS OF THE AMERICAN ABOKIGIXES. 4()7

cranial perforation is large and oval, somewhat less than half the
breadth of the shaft at the same point. The deltoid ridge is typically
prominent. The bone itself is slender and not in any way thickened
or distorted. It measnres: — greatest length, 130 mm.; antero-
posterior diameter of head, 31; transverse diameter of head, 25;
transverse diameter of distal end, 25.5; width of distal articular
surface, 17. It is thus about three quarters the length of the humerus
in the Larger or Common Indian Dog, proportionally slender, yet
considerably longer than that of the Techichi. What is undoubtedly
the radius of the same dog, measures 129 mm. in greatest length;
14.5 in diameter at the proximal and 19 at the distal end. A femur,
possibly of the same specimen measures: — greatest length, 136 mm.;
greatest transverse width of distal end, 25. It is thus slightly longer
than the humerus, in the normal proportion. The limb-bones indi-
cate a dog about the stature of a terrier or a basset-hound.

Among many isolated lower jaws from Maine shell-heaps are some
in which the carnassial tooth is noticeably narrow and intermediate
in size between that of the typical Short-legged Dog and the Larger
or Common Indian Dog. These probably represent cross-bred
animals as Loomis and Young have suggested.

Uses. — These smaller dogs were apparently the faniiliar household
pets or hunting companions of the Indians of forested country or of
the canoe-using tribes. They were too small to be of service as pack-
animals with travois or pannier, and hence seem not to have been
much in favor with the Plains Indians, whose main subsistence was the
Bison for the hunting of which, dogs were unnecessary. Suckley
(1860) particularly mentions that they were kept more as a "play-
mate for the children and a pet for the women" among the tribes of
the Columbia River. Moreover, a small dog is a better companion
in a canoe than a larger clumsy animal.

Richardson says of the Short-legged Dog, that it was used in the
chase, was very active and agile at jumping. It was perhaps a dog
of this type that was used in hunting the beaver. George Bird Grin-
nell (Forest and stream, 1897, 49, p. 382) writes that the Cheyenne
Indians, before their intercourse with whites, hunted the Beaver with
dogs, by breaking the dam and thus exposing the beaver houses and
their underwater entrance. "The dogs which were small enough to
enter this hole, and yet were pretty good sized animals, went into the
hole " and worried the beaver till it followed the dog out, when an
Indian waiting outside, clubbed the beaver to death. Le Jeune, in
his Relation de ce qui c'est passe en la Nouvelle France [Quebec]

468 bulletin: museum of comparative zoology.

en I'anne 1633 (Jesuit relations, 1897, 5, p. 165) mentions this use
of dogs in Beaver hunting; "sometimes when the dogs encounter the
Beaver outside its house, they pursue and take it easily; I have never
seen this chase, but have been told of it; and the savages highly value
a dog which scents and runs down this animal." Le Jeune speaks of
the familiarity of the Indian dogs, that in winter they are unable to
sleep outside and come into the cabins, lying and walking over the
inmates. Elsewhere he speaks of giving food to a 'petit chien,' but
adds that " the savages do not throw to the dogs the bones of female
Beavers and Porcupines, — at least certain specified bones .... yet
they make a thousand exceptions to this rule, for it does not matter
if the vertebrae or rump of these animals be given to the dogs, but the
rest must be thrown into the fire."

Testimony of early travellers is somewhat conflicting as to the
eating of their dogs by the Indians. Le Jeune states that " in the
famine which we endured, our savages would not eat their dogs,
because they said that, if the dog was killed to be eaten, a man would
be killed by blows from an axe." On other occasions, however, such
scruples were not observed. Thus Father Rasles, in a letter written
to his brother in 1716, from Narantsook, forty miles up the Kennebec
River, Maine, says that at the news of the French and English War,
the Indian young men were ordered by the older Indians to kill dogs
for the purpose of making the war-feast (Jesuit relations, 1897, 67,
p. 203) — possibly here with a view to sending their dogs on before,
should death overtake their masters. Feasts of dog-flesh seem to
have been commoner among the Indians of the West and South, and
Fremont in his narrative of his explorations (1845, p. 42) recounts
being invited, as a mark of honor, to a dog-feast. " The dog was in a
large pot over the fire, in the middle of the lodge, and immediately
on our arrival was dished up in large wooden bowls, one of which was
handed to each. The flesh appeared very glutinous, with something
of the flavor and appearance of mutton. Feeling something move
behind me, I looked round, and found that I had taken my seat among
a litter of fat young puppies."

Harmon, writing in 1820, after nineteen years spent in traxel
through the Northwest from Montreal to the Pacific, speaks of the
smaller dog used in hunting, and a larger dog as well. The latter is
rank and not good eating like the former, of whose flesh the Indians
and French Canadian voyagcurs were very fond.

In the New England shell-heaps, the dog-remains occur either as
burials — the entire skeleton undisturbed — or as scattered portions.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 469

as if the bones had been thrown out after the flesh was eaten. There
seems, however, to be Httle or no evidence that the bones w^ere cracked
for marrow.

The Jesuit father Biard in 1616, mentions dogs, kettles, and axes as
among the presents given by a young Indian to the father of his
intended bride in payment for her. Among other customs of the
Indians of Arcadia, he recounts that at a funeral, dogs are presented
the dying man, as well as skins, arrows, and so forth. The dogs are
then killed in order to send them on before him to the other world,
and their flesh is later eaten by the people (Jesuit relations, 1896,
3, p. 101).

Clallam-Indian Dog.
Plate 4, fig. 1.

1840. Canis laniger Hamilton Smith, Jardine's Nat. library. Mammalia-

10, p. 134.
1867. Canis domesHcus, camtschatkensis longipilis Fitzinger, Sitzb. K. akad.

wiss. Wien, 56, pt. 1, p. 406.

Characters. — A medium-sized dog, with erect ears, and bushy tail.
Hair rather thick and woolly; white, or perhaps brown and black.

Distribution. — Formerly found among the coast Indians of the
Puget Sound region and Vancouver Island. Lord (1866, 2, chap. 11)
asserts that these dogs seem to have fii'st been kept by the Chinook
Indians, once very numerous near the mouth of the Columbia River,
and were thence carried to Puget Sound and Nainimo. The source of
this information is not given, but it is worth remarking that Lewis
and Clark make no mention of the breed on the Columbia. Van-
couver found them near the then Port Orchard, and apparently at
least as far up the Sound as ■- Admiralty Inlet. Hamilton Smith
implies that they were to be found at Nootka Sound on the west
coast of Vancouver Island.

Descriptions. — The earliest account of this dog is that by the navi-
gator, Vancouver (1798, 1, p. 266). In May, 1792, while at Port
Orchard, Puget Sound, he writes: —

" The dogs belonging to this tribe of Indians [at Port Orchard] Avere
numerous, and much resembled those of Pomerania, though in general
somewhat larger. They were all shorn as close to the skin as sheep are
in England; and so compact were their fleeces, that large portions

470 bulletin: museum of comparative zoology.

could l)e lifted up l)y a corner without causing any separation. They
were composed of a mixture of a coarse kind of wool, with \-ery fine
long hair, capable of being spun into yarn. This gave me reason to
believe, that their woollen clothing might in part be composed of this
material mixed with a finer kind of wool from some other animal, as
their garments were all too fine to be manufactured from the coarse
coating of the dog alone. The abundance of these garments amongst
the few people we met wnth, indicates the animal from whence the
raw material is procured, to be very common in this neighborhood;
but as the}' have no one domesticated excepting the dog, their supply
of wool for their clothing can only l)e obtained by hunting the wild
creature that produces it; of which we could not obtain the least
information." Elsewhere he mentions a deer "they had killed on the
island, and from the numlier of persons that came from thence, the
major part of the remaining inhabitants of the \'illage, with a great
number of their dogs, seemed to have been engaged in the chase,"
this near Admiralty Inlet. Farther up Puget Island, 48° 2|'N, 237°
57|^'W, at a large village " they were met by upwards of two hundred
[Indians], some in their canoes with their families, and others walking
along the shore, attended by about forty dogs in a dro\e, shorn close
to the skin like sheep [this in June]" {Ibid., p. 284).

Hamilton Smith (1840) who, in addition to Vancou\er's account,
had information from an Indian who had resided two years at
Nootka, speaks of it as a large dog, " with pointed upright ears, docile,
but chiefly \aluable on account of the immense load of fur it bears on
the back, of white, and brown, and black colovirs, but having the
woolly proportion so great and fine, that it may well be called a fleece."

Notwithstanding Smith's assertion as to the " brown and black
colours" of this dog, it is not at all certain that this was the usual case.
Suckley (1860, p. 112) says positively that "all the Clallam dogs
that I saw were pure white; but they have the sharp nose, pointed
ear, and hang-dog, thicAash appearance of other Indian dogs." Gibbs
also {Ibid.) mentions their whiteness only, and adds that the very
soft hair is sheared like the wool of sheep, and made into blankets,
though at that time, 1860, it was "generally intermixed with the
ravellings of old English blankets to facilitate twisting with [?into]
yarn."

Lord (1866) further remarks that this white, long-haired dog was
kept by only a few coast tribes near Vancouver. The dogs were
confined "on islands to prevent their extending or escaping," and it
differed "in cAery specific detail from all* the other breeds of dogs

ALLKX: DO(;s OF THE AMERICAN ABORIGtXES. 471

belonging to either coast or inland Indians." He supposes it to be of
Japanese origin, recalling the long-haired Japanese Lap-dog, which
however, seems remote enough in other characters. Lord adds that
in the manufacture of rugs from the hair of this dog, the Indians often
added the wool of the Mountain Goat, or duck feathers, or wild hemp.
They dyed the hair as well. He obtained several of these blankets
along the coast for the British Museum. Newcombe (1909, p. 50)
gives a further account of the method of making yarn from the hair,
which he says, was remo\ed from the dried skin of the dog with
knives or pulled out after moistening the hide and "sweating" the
liair To loosen the roots. The wool was then made into loose threads
by rolling. V\h\\ the introduction of Hudson's Bay Company
blankets this industry has ceased aufl the dog was practically extinct
at the time of his writing.

As to the origin or affinities of this breed, little can be said. Some
writers have classed it with the Siberian and Eskimo dogs, but it is
likely that it was a breed of the larger type of Indian dog. The dis-
inclination to take to water, made use of by the Indians to confine
the animals to islands, is a trait shared by the Eskimo Dog. The
precaution was possibly taken in order to prevent crossing with other
breeds of Indian Dogs.

Windle and Humphreys (1890) in their table of cranial proportions
of Eskimo Dogs, include those of a Nootka Dog in the British Museum.
It is not clear, howe\er, if it ^\as from a dog of the breed under con-
sideration, and as no actual dimensions are gi\en, the figures are not
comparal^le with other direct measurements.

I am indebted to Mr. ('. T. C'urrelly, Curator of the Royal On-
tario Museum of Archaeology at Toronto, for a photograph (Plate 4,
fig. 1) of the unique painting made at A'ictoria, B. C, in 1846, by
Paul Kane and now at that ^Museum. In the foreground is one of the
white woolly dogs in question, its apparently erect ears nearly hidden
in the long hair of the head. Xearl)y an Indian woman is weaving
a blanket, no doubt from yarn made of dogs' hair, a ball of which
another woman in the background is spinning. The use of dogs'
hair in making blankets is not confined to the Clallams. The ancient
Zunis appear to have made similar use of it; and Bannister Q869)
mentions an Indian blanket from Mackenzie River, woven of dogs'
hair. The natives of New Zealand regularly employed dogs' hair
for braiding and ornament.

472 bulletin: museum of comparative zoology.

Inca Dog.
Plate 9.

1844. Canis ingae Tschudi, Unters. liber die fauna Peruana. Therologie,

p. 13, 249.
1885. Canis ingae peciiarms Nehring, Sitzb. Gesellsch. naturf. freunde

Berlin, p. 5-13.

Characters. — This is the hirger dog of the ancient Peruvians. It
was about the size of a small Collie, l)ut more heavily proportioned.
Tschudi describes it as having the head small, snout rather sharply
pointed, upper lip not cleft; ears erect, triangular, small; body short
and strong, squarely built ("untersetzt"), legs rather short; tail
about two thirds the length of body, fully haired and curled fonvard.
Pelage rough, long, and thick; color dark ochre-yellow with dark
wavy shadings; belly and inner side of limbs somewhat brighter than
the ground color of the back. No light spots above the eyes.

The skull is heavy in proportion to its size, with a narrow rostrum.
The brain-case is rugose for the attachment of muscles, yet the tem-
poral muscles, even in old dogs seem to little more than meet medially,
so that at most only a low sagittal crest is formed in old aninials
except at the extreme occiput, where it is contrastingly marked, form-
ing a high knife-edge on the median line of the interparietal. The
palate shows a strong thickening at its posterior end, forming two low
ridges one on each side between the last molar and the posterior narial
opening.

DistrihutioiL — The former distribution of this breed has not been
definitely traced. ]\Iummified remains are known from Ancon,
Peru, and from various sites that have been excavated in that country.
In Tschudi 's time it appeared to be confined to the upland tribes of
Indians. Of this type were all the mummies and skulls of dogs
found by him in the ancient graves among the Sierras. It probably
was kept by the Indians of northwestern Argentina as well.

Nomenclature. — Tschudi in 1844, was apparently the first to name
this as a distinct breed of dog, Canis inc/ac. Forty years later Nehring
in writing of the dog-mummies from the ancient necropolis of Ancon,
referred it to a collie-like type with the combination, Canis ingav
pccuarins. It is, however, very different cranially and otherwise
from the Collie.

Measurnnrnfs. — The largest Inca Dog among those from Ancon

ALLEN: DOGS OF THE AMERICAN ABORIGINES.

473

studied by Xehring (18S4a) was smaller than a Sheep-dog, with a skull
about 172 mm. long, humerus 147, ulna 172, radius 140. A smaller
one had a skull length of 165, head and body 060, tail including hair
240, humerus 130. In the lower jaws the first premolar was fre-
quently missing.

The following table gives measurements of the six largest skulls
among a series of nine belonging to the U. S. N. M.

Measurements of the Skulls

172.888

172,850

176.310

172.858

170,386

176,.309

Length, (occiput to anterior base

of incisors)

155

164

160

163

172

178

Basal length

139

145

146

144

151

159

Palatal length

Orbit to tip of premaxillary

Upper tooth-row

—

(alveoli)

Front of canine to back of molar-

(crowns)

—

Front of canine to back of molar-

(alveoli)

Length of premolar* (crown)

16.5

17.5

(alveolus)

16 .

16.5

" " -molars' ~- (crowns)

—

« " (alveoli)

15.5

16.5

Zygomatic width

108

107

Breadth of occipital condyles

Median length of nasals

48.5

—

Remarks. — Writing about 1844, Tschudi describes the chief char-
acteristics of this dog as treachery and mischievousness. Every
Indian hut and shepherd of the. Sierra and puna had several. They
seemed to show a special antipathy toward white people. A Euro-
pean traveller approaching an Indian hut on horseback woiUd be beset
by these dogs springing up against his horse to bite his legs. They
are courageous, and fight an enemy with determination, dragging
themselves to the attack even when mortally wounded. The Indians
train them to track and capture tinamous.

In their great work on the Necropolis of Ancon, Reiss and Stiibel
include a brief chapter by Nehring (1884b) on the mummified remains
of dogs discovered there. Some of these are figured and show a pale

474 bulletin: museum of comparative zoology.

yellowish coloring with darker areas. In a more extensive article
Nehring (lS84a) gives a particular account of the dogs of Ancon.
He first transcribes passages from Garcilasso de la Vega to show that
the Incas had dogs previous to the Spanish conquest, and that the
dog entered into certain religious rites of the Incas. A munmiified
dog is described as having thick hair, shorter, however, on head and
feet, thickest on neck and breast forming a kind of mane.~ The color
was yellow, clear or soiled in places, with irregular brown-shaded areas.
The tail was thick and bushy, wolf-like, also yellow. The ears of
most of the specimens seemed to ha^■e been clipped. He suggests the
North American Wolf or Coyote as the original source of the Inca
dogs, but there seems no ground for the selection of either as an
immediate ancestor.

More recently, Eaton (1916, p. 25) has recorded the discovery of
dog-mummies with pre-Columbian burials at Machu Picchvi, Peru.
He adds that " dogs of this general type, though usually a little smaller
than those figured in Reiss and Stiibel's Necropolis of Ancon, were
frequently seen in the parts of the Cordillera that I visited, and these
animals may be largely derived from the ancient stock. . . The
modern Indian dogs of this ancient type are very wolf-like and mani-
fest a most inconvenient fear of the camera." He suggests the obvious
possibility of present-day mixture with breeds imported from
Europe, and gives a reproduction (p. 50, fig. 47) of a photograph
showing dimly an Indian with his dog.

The fine series of Peruvian dog-skulls in the U. S. X. ^I. contains
nine that show complete gradation in size between the smallest (which
I have considered more or less typical of the Techichi) and the largest
which represents the Inca Dog. Since these skulls are more or less
comparable as to age, it seems likely that the gradation in size is due
to free interbreeding of the two sorts of dogs. The largest skull of
the series (U. S. N. M. 176,309, of which the measurements have been
given) is almost precisely matched by the skull of a Common Indian
Dog from Peel River, Arctic America, collected by Robert Kenni-
cott about 1860 (U. S. N. M. 6,219). The only obvious differences
are that the palate of the Inca Dog shows the peculiar thickened ridges
at the posterior end and is much narrower across the occipital con-
dyles. The latter characteristic is shared by the other dog-skulls
from Peru in contrast with the northern dogs, and is no doubt among
the latter a result of their use as sledge-dogs, for the greater develop-
ment of the neck and chest muscles in hauling might well enough
demand a broader support from the skull. This general similarity

ALLEN: DOGS Of^ THE AMERICAN ABORIGINES. -1:75

of skull and skeletal proportions probably indicates a closer relation-
ship with the larger Indian dogs of northern North America, than with
the Wolf or Coyote as Nehring has suggested.

What may be feral dogs of this breed are said to be found in the
Island of Juan Fernandez, off Peru. According to Ermel (1889, p. 53)
they are the native Araucarian dogs, shaggy-coated, of medium size,
and very powerful. Semitamed ones are sometimes used there in
Imnting the feral goats.

Ihering (1913) has recorded the discovery of an entire skeleton of a
dog at Hualfin, Salta Province, in northwestern Argentina. Its
skull measurements, as recorded by this author, correspond well with
the larger of those above given, and his identification of the specimen
as an Inca Dog is probably correct.

Long-haired Inca Dog.

Characters. — Apparently similar to the Inca Dog, but with longer
coat.

Distrihution. — Peru and probably coastwise to parts of Chile.

Notes. — In his Bibliography of the tribes of Tierra del Fuego and
adjacent territories. Cooper (1917, p. 44) mentions "a breed of long-
haired shaggy dogs" which was fonnerly raised among some of the
Chonos Indians north of the Taitao Peninsula, Chile, about Lat. 45°
South. Nothing is known about these dogs except the statements of
Goicueta and Del Techo, based perhaps on independent testimony.
It is assumed that this breed was of native origin since at that early
date (about 1553) it is rather unlikely that such dogs would have
been obtained from Europeans. Possibly they were derived from the
larger collie-like type of Inca dog anciently found among the Peruvians
(Eaton, 1916, p. 49). From the hair of these dogs, the Chonos made
short mantles that covered the -shoulders and upper part of the trunk.
According to Cooper, the information of Goicueta is based on the rela-
tion of Cortes Hojea's expedition of 1553-54, when he commanded
one of the vessels under Ulloa, and possibly also furnished one of the
sources for Del Techo's account. The latter was a Jesuit missionary
who wrote in 1G73 concerning the labors of his brethren among the
Chonos of the Guaitecas Islands.

Referable to this breed is probably the long-haired dog described
by Nehring (18S7a) from a well-preserved mummy found in the course
of excavations at Ancon, Peru. It was found wrapped in cloth of

476 bulletin: museum of comparative zoology.

tree-wool, its head and feet tied together. In the size of its skull
and leg-bones it was said to be like the ordinary Inca Dog of the collie-
like type, but clothed with unusually long hair, especially on the feet
and tail. The hair is described as of a dull yellow. This dog must
have been very similar to the Long-haired Pueblo Dog previously
mentioned as discovered by Messrs. Guernsey and Kidder in excava-
tions at Marsh Pass, Arizona.

Patagonian Dog.

Characters. — A medium-sized dog, as big as a large Foxhound,
coat usually short and wiry, or longer and of softer texture; ears
short and erect; color dark, more or less uniform, rarely spotted;
dark brownish black, dark tan, or occasionally black; tail bushy.
General appearance like a small Wolf.

Distrilmfion. — Found among the Foot Indians of the eastern parts
of Tierra del Fuego, northward into Patagonia, the northwestward
limits of distribution not clearly known.

Remarks. — Hamilton Smith (1840, p. 213) quotes a letter from
Captain Fitzroy of the Beagle, that the Patagonian Dog is strong,
about the size of a large Foxhound, coat short and wiry, though
sometimes soft and long, like that of a Newfoundland Dog. In color
it is dark, nearly unifonn, rarely spotted. It is wolfish in appearance,
somewhat resembles the Shepherd Dog, will growl and bark loudly.

It is doubtless a dog of this breed that is meant by Furlong in his
statement that of the two types of dogs found among the Onas of
Tierra del Fuego, one is like a \Yolf.

Cunningham (1871, p. 307) mentions that while near Gente Grande
Bay, Sandy Point, in the Strait of Magellan, three dogs wandered
about in the neighborhood of his landing party, " barking and howling
dismally. The first was very much like a fox in size and general
appearance, and of a reddish-gray colour; the second had a piebald
smooth coat, with drooping ears; while the third was clothed with long
dark brownish-black hair, had erect ears, and presented a marked
resemblance to a small wolf." The first was probably a Fuegian Dog,
obtained through intercourse with tribes of the western part of the
Magellanic Archipelago ; the second was possibly a mongrel European
dog; the last perhaps a Patagonian Dog.

Of this animal, Spegazzini (1882, p. 176) writes that it differs greatly
from the Fuegian Dogs of the Canoe Indians, "y para mi serian 6

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 477

cruza 6 descendientes directos del lobo-colorado 6 gran zorro-colo-
rado." It is difficult, however, to see any ground for deri\ing it
from the peculiar Pampean Wolf. It is much larger than the Fuegian
Dog, and is described by Spegazzini as tall, slenderly built, with fierce
eyes; long-haired and bushy-tailed; the color prevailingly dark tan,
but occasionally black; rather silent, not barking though gi\ing voice
to melancholy howls.

Fitzroy (see Hamilton Smith, 1840, p. 215) particularly describes a
dog seen near the Strait of LeMaire. No temptation would induce its
master to part with it. It was the size of a large setter, with a " wolf-
ish appearance about the head, and looked extremely savage. Behind
the shoulders it was quite smooth and short-haired, but from the
shoulders forward it had thick rough hair," giving it a lion-like ap-
pearance, " of a dark grey colour, lighter beneath, and white on the
belly and breast; the ears were short but pointed, the tail, smooth
and tapering;" the fore c^uarters very strong but the hinder appearing
weaker. The short-haired tail seems unnatural for a Patagonian Dog,
and may have been evidence of a strain of blood from a European
source.

The eastern Fuegians or Onas, are considered by ethnologists to be
derivatives of the Patagonians, and no doubt originally had these
dogs from their mainland relatives, or brought them at the time when
thev colonized the Fuegian countr\'.

It is unfortunate that no bones or figures of the Patagonian Dog
are available for comparison. Ihering (1913) has, however, recorded
the skull of a prehistoric dog from Amaicha, Tucuman province,
northwestern Argentina, which may represent it, and at the same
time indicate nearly its northern range. This skull was 190 mm. in
total (?occipitorostral) length, the upper fourth premolar 19 mm.,
the combined upper molars 20 mm., hence a somewhat larger breed
than the Inca Dog.

The native Patagonian Dog is not to be confused with the dogs
introduced by Europeans, that have since become feral on the pampas
of southern South America. These, according to various writers
(Rengger, 1830; Hamilton Smith, 1840; Rasse, 1879) are mongrel of
several breeds, notably one like the Great Dane. They are said to
go in troops and to make burrows in which to shelter their young.
This burrowing habit has been noticed in case of other feral dogs.
Thus Coues (1876) records the case of a brindled cur that became feral,
and took up its habitation in a burrow on the open prairie, near
Cheyenne, Wyoming, and in this den had a litter of fi^•e puppies.

478 bulletin: museum of comparative zoology.

Fitzinger (1867, p. 397) applies to the feral Pampean Dog the Latin
combination " Canis domesiicus, pyrcnaicus alco" (!) and briefly states
that it is probably a hybrid between the Pyrenian Dog and the Bull-
dog. Hamilton Smith (1840) had previously described it under the
Latin name Canis campivagus.

As to the origin of the Patagonian Dog, there is little satisfactory
evidence, but it may be assumed to be a derivative of the same stock
as the Inca Dog. The tooth measurements of the skull recorded by
von Ihering (1913), cf. p. 477, accord very nearly with those of the
largest Inca Dog of our table (p. 473), though even larger.

Mexican Hairless Dog; Xoloitzcuintli.
Plate 2; Plate 3, %. 2.

1651. Lupus mexicanus Recchi and Lynceus, Rerum medicarum Novae

Hispaniae thesaurus, p. 479, fig.
1766. Canis 7nexica7ius Linne, Syst. nat., ed. 12, 1, pt. 1, p. 60, (based on

Recchi and Lynceus).
1788. Canis familiaris aegyptius GmeUn, Linne's Syst. nat., ed. 13, 1, pt. 1,

p. 68 (in part).
Canis familiaris orthoius xoloitzcuintli Reichenbach, Naturg. raubth.,

p. 150.
1821. Canis nudxis Schinz, Cuv. thierreichs, 1, p. 218.
1827. Canis familiaris caraibaeus Lesson, Man. mammalogie, p. 163.
1844. Canis caraibicus Tschudi, Fauna Peruana, Therologie, p. 249.
1887. Dysodus gibbus Cope, Amer. nat., 21, p. 1126.

Characters. — A dog of medium-size, rather heavily built, and
long-bodied in proportion to its height; ears large and erect; tail
thick, drooping or carried nearly straight behind; hair nearly absent
except for a few coarse vibrissae and generally a sparse coating on the
tail, particularly near the tip; sometimes a tuft on the crown. The
skin is usually pigmented, a slaty gray, or reddish gray, paler in
the bends of the legs; sometimes blotched with wdiite.

Distribution. — This race seems to have been native among the
peoples of Central and South America from Chihuahua perhaps con-
tinuously southward, to the Peruvian lowlands, and in some of the
Greater Antilles; it may also have been indigenous among the In-
dians of Paraguay.

History.^ The first account of the Mexican Hairless Dog by a

ALLEN: DOGS OF THE AMERICAN ABORIGLNES. 479

European, seems to be that of PVaneisco Hernandez, who lived
between the years 1514 and 1578. His Historia AniniaHum et Minera-
lium Novae Hispaniae, is printed on 96 folio pages as part of Recchi
and Lyneeus's Rerinn ^Vledicariim Novae Hispaniae Thesaurus,
1651, which was apparently intended as a monographic elaboration
of Hernandez's work. This writer brought l)ack an account of three
sorts of dogs, which were in his day kept by the native Mexicans.
The first of these he had himself seen, but the two others he had
neither seen, nor known of their having been ])r()ught to Europe.
This first sort he states, is called the Xoloytzcuinili and is larger than
the others, exceeding three feet in body length, but with the peculi-
arity of having no hairy covering, yet with a soft skin, spotted with
fulvous and slate color. (" Primus Xoloytzcuintli \ocatus alios
corporis vincit magnitudine, c^uae tres plerum; excedit cubitos, .sed
habet peculiare nuUis pilis tegi, verum molli tantum, ac depili cuti,
fuluo atque Cyaneo colore maculata."). The two other sorts of <logs
were the hump-backed or Michuacan dog and the Techichi, elsewhere
discussed. The XoloyizcuinUi of Hernandez is clearly the Hairless
Dog, and a most elaborate account of the animal is given by Recchi
and Lynceus (1651, p. 47C ft'.) with a fairly recognizable figure (Plate 2,
fig. 1). These authors apparently had an actual specimen, possibly
one brought alive to Europe; at all events they describe its appearance
as fierce and wolf -like, with a few bristly hairs about the mouth, the
mammae ten as in the wolf and dog, and the vertebrae of the same
number as in a dog-skeleton with which they compared it, namely
seven cervicals, thirteen dorsals, seven lumbosacrals, seventeen caud-
als. They name the animal Lupus mc.vicamis in contradistinction
to their Alco or Cauls ituwicana, which was probably a Raccoon.
This name appears in zoiilogical nomenclature in the twelfth edition
of Linne's Systema naturae under the genus Canis. The diagnosis,
evidently based on the figure and description just noticed, reads:
" C. Cauda deflexa lae\i, corpore cinereo fasciis fuscis maculisque
fulvis variegata"; the habitat is given as the warmer parts of Mexico.
Linne's first reference is to Brisson, whose description — " Canis
cinereus, maculis fulvis variegatus" — is clearly from the same
source. Hitherto Linne's Canis vw.vicanus has been regarded as
applying to the wolf of Southern ^Mexico, }>ut no true wolf is known
from that part of the country. Miller (1912a) seems to have been the
first to question the propriety of using the name for a wolf, but leaves
the matter unsettled, saying that according to E. W. Nelson, "the
wolf of the southern end of the Mexican tableland became extinct

480 bulletin: museum of comparative zoology.

about fifty years ago" (1860). Some other name must therefore be
apphed to this wolf if it ever be shown to be distinct.

The above accounts by Hernandez and by Recchi and Lynceus are
the basis of most of the earher references to the Mexican Hairless Dog.
Lesson, in 1827, however, redescribed it under the name caraibacu.s,
and Gmelin, earlier, 1788, had considered it the same as the Turkish
or Egyptian Hairless Dog, under the name Cam's f. acgnpiivs: this
however, is a hairless variety of another breed.

Noics. — The former distribution of this remarkable dog is now
hardly tracealile with certainty except in a general way, but it was
kept by the Mexicans of Chihuahua and southward, as well as by
the natiA-es of Peru, more especially those of the lower altitudes.
According to Seler (1890) the INIexicans wrapped these dogs in cloths
at night as a protection against cold. Some were not naturally
hairless, but were rubbed with turpentine from early youth, causing
the hair to fall out. On the other hand, dogs naturally hairless were
raised, as at the pueblos Teotlixco and Tocilan. The Zapotec and
jVIaya languages have separate words for the hairless dog. The term
a'oloifzcuhiili is said to signify the monstrous dog. Patrick Browne
(1789, p. 4S.()) writing of the natural history of Jamaica, mentions the
Indian dog as " Canis pilis carens, minor," a creature "frequent
among the Jews and nctiwes'^ in that island; he describes it as "gen-
erally about the size of a cur-dog with a rough skin, which looks like
the hide of a hog." There is nothing to indicate, howe\-er, that the
breed was common in the West Indies.

In Pern, Tschudi (1844, p. 249) observed this dog mainly on the
coast, since its lack of a hairy coat made it unable to withstand the
cold of the higlier altitudes of the interior except in the warm valleys,
and then only if carefully protected. He describes it as slaty gray
or reddish gray, sometimes spotted, and says it is voiceless. He is
probably mistaken, however, in supposing these were the dogs found
by ("olumbus among the Lucayans. Nearly twenty years prcA'iously,
Lesson had seen the Hairless Dog in numbers at Payta, Peru.

According to Rengger (1830), a hairless dog, possibly identical with
the ^lexican Hairless Dog, was indigenous among the Indians of
Paraguay, Avho had a special word — yagua — for it. He describes it
as ha\ing a relatively small head, pointed snout, ears erect or only
their tips drooping foi'ward, rump fat, extremities fine, tail spindle-
shaped nnd iisuidly drooping. Some indi\iduals do not bark, but
howl only.

During the last hundred years, little attention seems to ha\e been

ALLEN: DOGS OF THE AMERICAN ABORIGLNES. 481

given to this breed, although lately it has been taken up by dog fan-
ciers. LeConte, in 1856, calls it the Comanche Dog, and says it is
common among the Indians of that tribe, but, " though some of these
dogs ha^■e been brought within the United States, we have no descrip-
tion of them." Packard (1885) mentions seeing one in his visit to
Mexico, but they were apparently uncommon. In a recent letter
from Mr. Arthur Stockdale, he states that in Mexico ( 'ity they are
now considered somewhat of a rarity, though said to be common in
Chihuahua, where however, little attention is paid them.

There is some evidence that they do not breed readily with normally
haired dogs, yet such crosses have been made, and curiously the result
seems to be that about 50% of the young are naked or practically so,
the other 50% fully haired. Stockdale (1917) records such a litter
consisting of two puppies, one hairless, the other normal. Kohn
(1911) records a mating of a Hairless Dog with a Fox-terrier, the four
offspring of which comprised two naked and two completely-haired
dogs. His microscopic study of the skin of the Hairless Dog indicates
that its character is that of a young embryo's, whence it may be that
the hairless character is merely the retention of the embryonic condi-
tion, just as the short-nosed skull of the Japanese I^ap-dog seems to
be a case of the retention of the embryonic proportions of the skull.

As to the origin of this breed, it is most likely a variant of the larger
type of Indian Dog, in which the hairlessness is due to a retention of
the emljryonic condition of the skin, precluding hair development,
just as the short-nosed Ijreeds of dogs are the result of the failure of
the facial bones to attain full growth.

I have unfortunately l)een unable to obtain skulls for comparison.

Small Indl\n Dog or Techichi.
Plate 10.

1788. Cains familiaris americatiws Gmelin, Linn6's Syst. nat., ed. 13, 1,

pt. 1, p. 66 (in part).
1792. Canis americanus ■plancus Kerr, Animal kingdom, 1, p. 136 (based on

the Techichi of Hernandez).

1840. fCanis alco Hamilton Smith, Jardine's Nat. library. Mammalia, 10,
p. 135, pi. 4, left-hand fig.

1841. ? Canis familiaris cayemietisis Blainville, Osteographie. Atlas, ]:>!. 7^
1867. Canis caraibaeus, hernandesii Fitzinger, Sitzb. K. akad. wiss., Wien,

56, pt. 1, p. 498.
1882. .'Canis ffibbus Duges, La naturaleza, 5, p. 14, fig. 1-3. .,,.

482 bulletin: museum of comparative ZOOLOGV-

Charactcrs. — X small, light-limbed dog, of rather slender propor-
tions, narrow delicate head, fine muzzle, erect ears, well-developed
tail, which may have been close-haired. Colors black, black and
white, or perhaps brownish or yellowish.

Distribution. — This was perhaps the dog of fox-like appearance
noticed by many of the early explorers, yet it is difficult to indicate
the limits of its former distribution. On the Atlantic seaboard, among
the considerable quantity of skeletal remains exjimined, I have seen
nothing that could be referred to such a dog; yet Brereton, who
reached the Elizabeth Islands and coast of southern New England
with Gosnold in 1002, mentions "Dogs like Foxes, blacke and sharpe
nosed" among the "Commodities" seen there. In the famous
village site near Madisonville, southwestern Ohio, its bones occur
and there are in the Peabody Museum similar bones from the south-
west and Yucatan, believed ec^ually to be pre-Columbian. Among
the dog-skulls found with Peruvian burials the same type occurs, as
well as skulls intermediate between this and other dogs, and so proba-
bly representing mongrel individuals. Probably then this type of
dog was spread over at least the central and southwestern part of
North America and parts of northwestern South i\.merica.

Nomenclatuir. — This is assumed to be the Techichi of the earl\'
Spanish accounts of JVIexican dogs, though there is little doubt that
two different animals as well as more than one breed of dog were con-
fused under this title by the early writers and systematists. It is of
some importance, therefore, to examine their accounts carefully since
the case is somewhat complex and involves the identity of the AIco
of early writers. Both Gmelin and Kerr based their names on the
account of Recchi and Lynceus (1651, p. 46G), who in turn refer to
Hernandez's brief account (which they print), in the Historiae ani-
malium et mineralium Novae Hispaniae, page 7. Hernandez who died
in 1578, had visited Mexico, and in his enumeration of its animals
includes three sorts of native dogs. The first of these is unquestion-
ably the Mexican Hairless Dog, and as'he himself states, was the only
one he saw personally ("caeteros vero neque conspexeram, neque
adhuc eo[z. c. ad Europam] delatos puto").

His account of the two other dogs is important and reads: —
" Secundus Melitensibus canibus similis est, candido, nigro, ac fuluo
colore varius, sed giberosus, gratusque iucunda quadam deformitate,
ac capite velut ab humeris edito, quem Michuacanensein abora vnde
est oriundus vocare solent. Tertius vero nuncupatus Techichi,
Catulis similis est nostratibus, Indis edulis, tristi aspectu, ac caetera

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 488

vulgaribus similis. Atque haec de canibus Nouae Hispaniae breuiter
dicta sunto." Translated freely, "The second is like the Maltese
dogs, in color varied with white, black, and fulvous, but it is hump-
backed and prized for this pleasing deformity, and a head that appears
to grow from the shoulders. It is called the Michuacan flog from
the place where it is native. The third sort of dog, however, is called
Techichi, and is like our Spaniels, but of sad countenance, though in
other respects like ordinary dogs. It is eaten by the Indians. This
then is briefly what I have to say of the dogs of Mexico." The
Techichi apparently was in no wise peculiar as a small dog. The
Michuacan animal, however, was hump-backed, without conspicuous
neck, its colors white, black, and fulvous, 'Sarins." In their elabo-
ration of Hernandez's account, Recchi and Lynceus (1651, p. 466)
fail to distinguish between these two supposed dogs; at all events
their figure (Plate 3, fig. 1) and description deal altogether with the
hump-backed animal, of which they seem to ha\'e had some knowl-
edge or probably a preserved specimen. They figure a female under
the name ' Canis Mexicana ' and the Mexican name Ytzcuintcporzotli,
the firSt half of which signifies 'dog.' Buff on, and later Gmelin,
likewise failed to distinguish between Hernandez's second and third
sorts of dogs, and the latter author in 1788, combined the two under
the name Avicricanus, with a brief diagnosis based on the figure of
Recchi and Lynceus, viz., "magnitudine t [i. c. of the breed inelitaeus],
capite parvo, auribus pendulis, dorso curvato, cauda brevi." Under
this name, Gmelin included: a. Ytzcuintcporzotli, or the Canis mexi-
cana of Recchi and Ljoiceus and b. Techichi of Hernandez. Obviously
the diagnosis applies to the hump-backed animal only, to which
Buffon had already applied the native name Alco, following Recchi
and Lynceus. This name appears to have been of doubtful applica-
tion to the common dog, but was used at times by later writers to
indicate the small native dog of Peru and Mexico. Kerr (1792, p. 136)
endeavors to improve on Gmelin by distinguishing with Latin names
the two varieties of the latter's Canis amcricamis. He first trans-
scribes the description and then distinguishes: "a. Fat Alco. —
Canis aniericanus obesus" and " b. Techichi. — Canis americanus
planeus," with descriptive accounts from Hernandez and his elobora-
tors, corresponding to Gmelin's "a" and "b."

What then was this Alco? A study of Recchi and Lynceus's
figure (Plate 3, fig. 1) and description seem to indicate clearly that they
had in mind a Raccoon. They describe its nose, forehead, and eye-
brows as white, these markings evidently delimiting the dark face,

484 bulletin: museum of comparative zoology.

while the pecuh'ar and characteristic upward slope of the back in the
live animal is thus described: "Dorsum cameli instar gibbosum, post
coUuni subito ad pectus accliue, sed coxas versus decline." The tail
is said to be short, barely reaching the heel, the mammae six in num-
ber. They further note its \ery fat belly, beautifully covered with
thick black hair Aaried with spots; feet and shanks whitish, claws
strongly exserted. These characteristics recall the Raccoon more
than an\' other animal. There are, however, eight mammae in this
animal, and the ears are not pendulous as described, biit these dis-
crepancies may be due to inaccuracy of observation, or the condition
of the specimen (perhaps a preserved hide) which the authors seem to
have had. The account quoted from Acosta (1590, p. 277) doubtless
refers to the same animal and not to a dog. This author, in his
Historia natural y moral de las Indias, writes: — " Verdaderos perros
no los aula en Indios, sino unos semejantes a perrillos, que los Indios
llamauan Alco: y por su semejana a los cjue ha sido lleuados de
Espaiia. tambien los Uaman Alco: y son tan amigos destos perrillos
que se quitaran el comer, por darselo : y quando van camino, los lleuan
consigo acuestas, o en el seno." (Of real dogs there are none' in the
Indies, save certain animals resembling little dogs, which the Indians
call Alco; and on account of their resemblance to our dogs brought
here from Spain, the Indians call these Alco as well: and so fond are
they of their little dogs that they deii}^ themselves of food in order to
give it to them; and when they go on a journey they carry the little
dogs with them on their shoulders or in their arms). The Raccoon
rather than a small dog seems to be indicated here, and the habit of
carrying them about on journeys would perhaps accoimt for the
present-day anomalous distribution of the small species of raccoon in
Central America (Panama) and in the islands of Cozumel, Guade-
loupe and New Pro^'idence. Acosta's story may also explain the
transference of the name Alco to small dogs, though Philippi (1886)
says this means dog in the Quichua tongue.

An early mention of the tame Raccoon is found in Hakluyt's Voy-
ages, in A relation of the commodities of Nova Hispania, and the
maners of the inhabitants, written by Henry Hawkes merchant,
which lived five yeeres in the sayd countrey, written in 1572. He
says: "Their dogs are all crooked backt, as many as are of the coun-
trey breed, and cannot run fast: their faces are like the face of a pig
or a hog, with sharpe noses."

If Gmelin's name amcricamis be admitted as applying to a Raccoon
it would antedate ^Yagler's name hernandezii (1831) for a Mexican

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 485

Raccoon. In view, however, of the uncertainty as to which form of
Raccoon it shoukl indicate, there seems to be no virtue in making-
such a change at present.

Later writers have tried to discover hving examples of the original
Aico with small success. Hamilton Smith (1840, p. 13.5, pi. 4, left-
hand fig.) describes as Canis alco, what he supposed to represent this
breed, from a stuffed specimen in an exhibition of Mexican ciu'iosities
made by W. Bullock, and said then to be in the Egyptian Hall (British
jSIuseum). He says of it: "That enterprising traveller descrilied it
as of the wild race; yet, from its appearance, we at first considered it
to be a Newfoundland puppy." The figure shows a small black and
white dog with rather full-haired tail, clumsy build, and ears laid
back. Of the mounted specimen, Hamilton Smith further writes: —
" It was small, with rather a large head; elongated occiput; full muzzle;
pendulous ears; having long soft hair on the body. In colour, it was
entirely white, excepting a large black spot covering each ear, and
part of the forehead and cheek, with a fulvous mark above each eye,
and another black spot on the rump; the tail was rather long, well
fringed, and white." This description, except for the pendulous ears
might apply well enough to the type of small dog here treated. How
much of its appearance was due to the taxidermist's efforts is, how-
ever, to be considered. It is even possible that it was after all only a
spaniel, which, except for its short ears, it seems to resemble.

What seems to have been a slightly deformed Indian Dog, is de-
scribed and figured by Duges (1882) as a Chihuahua Dog fa tenn that
is used by fanciers for a dwarf breed, with erect ears). From his
figure of the skull, it is evident that the animal was young. It was
apparently rather small, had but three lower premolars (the first
lacking), a rather heavy head, and long close-haired tail. The back
seems to have been unduly arched but the head is represented as
erect, and the posture quite different from that of a raccoon. The
color was blotched black and white. The ears were cropped, l)ut
were assvuned to have been erect. So far as can be judged from
Duges's account, this may have been a dog similar to the Techichi.
He, however, supposed it to represent the x\lco.

The confusion of names has been added to by Cope (1887) who
examined three skulls of the so called Chihuahua Dog. He found
a variable reduction in the number of teeth, correlated apparently
with the loss of hair. The premolars were reduced to § or f, while the
molars were §, ^, and § respectively. In all, the inner cusp of the
lower sectorial was lacking. On account of the reduced number of

486 bulletin: museum of comparative zoology.

molars, and this character of the sectorial, Cope refers this breed to
his genus Dysodus (Cope, 1879, lS79a) based on the Japanese Lap-
dog, adding that " the species may be called Dysodus gibbus," for
" the Chihuahua dog is the Canis gibbvs of Hernandez." The animal
to which Hernandez applied the adjective " (jibcrostis,''' however, was
with little doubt a Raccoon.

Skeletal Remains. — Among a great number of bones of Indian dogs
examined, from mounds, burials, or refuse deposits in various parts of
America, there occur skulls or fragments of jaws appertaining to a
wholly different type of dog from the large varieties just described.
The remains indicate a small light-limbed animal, with slender muzzle
abruptly narrowed in front of the third premolar. Although the
surface of the brain-case in adults is roughened for muscular attach-
ment the sagittal crest does not develop till old age. All the teeth
are small (upper carnassi^l 14-16.5 mm. in length), the nasals long,
and the skull normal, in that it seems not shortened or broadened in
any way, the teeth not crowded. A transverse line at the end of the
palate falls about through the middle of the second molar. These
dogs are probably the third variety of Hernandez, the Techichi or
Small Indian Dog. Several skulls, more or less imperfect, from the
Madisonville, Ohio, village site are referred to this breed, though
their measurements are a very little larger than those of more southern
specimens. They occur here together with bones of the large type of
Indian Dog. An imperfect cranium (M. C. Z. 7,123) collected many
years ago in McPherson's Cave, Virginia, by Lucien Carr, is apparently
in every respect similar to a skull of this type from Pecos, N. M.,
obtained by Dr. A. V. Kidder in the course of excavating a village site.
A similar but slightly smaller, though adult, skull from Pueblo exca-
vations in the southwest is practically the same, as is also a skull of
the Papago Indian Dog obtained by the late Dr. Edgar A. Mearns
at Sonoyta, Sonora, while on the Mexican Boundary Survey. It is
not fully adult, though of nearly mature dimensions. What seems to
be a dog of this type is represented in the Peabody Museum by a
cranium and hind leg-bones from Labna, Yucatan; the rostrum is
damaged and the teeth lost except the carnassial. The long slender
limb-bones are in strong contrast with the short thick bones of the
Short-nosed Indian Dog.

Turning now to South America, the Museum has a cranium from
Surinam, labeled: — Carib Indian Dog. It was received through the
Boston Society of Natural History from the Wyman Collection, and
was probably collected by Dr. F. W. Cragin, some fifty years ago

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 487

Though it has acquired the adult dentition, it is not old, and the
temporal ridges have not yet united to fonn a crest. A very similar
skull from French Guiana is figured by Blainville (1S41) under the
name Canisfamiliaris cayoinrnsis, by which he seems to have intended
to name the native dog.

I am indebted to Dr. \V. C. Farrabee for a photograph, (Plate 5,
fig. 2) which is assumed to illustrate this dog. It was secured by him
while studying the Macusi tribe in southern British Guiana, and
shows an old dog, and a puppy, accompanying a child of the tribe.
The larger dog has a narrow head, and erect ears, the tips of which
have been cropped, probably as a propitiation to evil spirits; the body
is short in proportion to the lean limbs, the tail (better seen in the
picture of the puppy) is long, upcurxing, and like the l)ody, short-
haired. Dr. Farrabee writes that these dogs "are small, yellow and
white, or brindle and white, and may be very much mixed with
European dogs." Of their ancestry, howe\er, there is no evidence,
though the erect ears and slender proportions faxor tlie supposition
that they retain a measure of their aboriginal character. The expres-
sion of the larger dog recalls the " tristi aspectu " of Hernandez's
description of the Techichi. It is not unlikely that the small dogs
found by the Jesuits among the Indians of the southern Antilles and
parts of Colombia and Central America may hnw been of the breed
here described.

Dr. Farrabee writes me further concerning son\e larger dogs whicli
he saw among the Wanoai tril)e "who occupy the Akarai Mountains,
northern Brazil to southern British Guiana. This tribe, on the
Brazil side had never seen white men before [his visit]. They have
the best dogs of all the tribes visited and they take the best care of
them. These dogs are noted among the tribes a month's journey
away. They keep the dogs tied on raised platforms and allow them
t^xercise morning and evening. The dogs are all black and white
and of good size." A small photograph of these dogs shows a hound-
like aspect and drooping ears. They are probably of European ori-
gin and perhaps the same as the dogs mentioned by Bancroft (1769,
p. 140) who says: " The Dogs of Guiana seem to be of a species between
the Hound and Land-Spaniel: their make is slender, their ears long
and pendulous, with a blunt nose, and large mouth: their bodies
are covered with long shaggy hair, generally of a fallow colour. The}-
pursue and start the Game by the scent."

I am indebted to J. Rodway, Esq., of the Museum at Georgetown,
British Guiana, for a brief note on the hunting-dog of the present-day

488 bulletin: museum of comparative zoology.

Indians of that country. He considers that it is of undoubted Euro-
pean origin, "has no particular characters," and "could be matched
in any lot of mongrels. It is generally rather small with a pointed
muzzle, foxy looking, and kept hungry to prevent laziness." The
"foxy" appearance is somewhat typical of the native breeds of smaller
Indian dogs, a result of the fine muzzle, ample erect ears, and drooping
tail, traits which seem still traceable among these mongrels of the
modern Guiana Indians.

Among a series of dog-skulls (l)elonging to the U. S. N. M.) from
ancient burials in Peru are two which in their small size and slender
proportions seem referable to the Techichi. Both are fully ai^ult,
with a well-developed sagittal crest on the interparietal, extending
forward in the larger skull on to the parietal suture. As will be seen
from the table of measvu'ements appended these skulls are a very
little larger, with slightly shorter nasals, as compared with the other
skulls whose dimensions are given. It is possible that this is due to
some admixture with the short-nosed breeds. Nevertheless the skulls
in question are quite different from the latter in their slender and
narrow outlines, and unshortened tooth-row.

No doul)t, flid we know the external characters of the dogs whose
skulls are here listed, it would be possible to recognize more than
one breed. Thus the Ohio individuals are a trifle larger in dimensions
than those of the Southwest and the Peruvian dogs again are a little
larger. Yet all are clearly of the same general type.

A comparison of the skulls and measurements of these specimens
with those of the Canis pahisfris oi Riitimeyer from the Swiss Lake-
Dwellings of late Neolithic to Bronze times in Europe, reveals a rather
close correspondence which is probably more than accidental, and
may even indicate a derivation from some common Asiatic stock at a
very early period. The type of small dog of the Swiss Lake-Dwellings
was one apparently of general distribution in southern Europe during
the Neolithic time, and AYoldrich (1886a) has identified it as far north
as Denmark in the kitchen-middens. It was apparently, on the
average, of wider zygomatic breadth, but otherwise its dimensions
corresponded very closely. This evidence favors the view that a dog
of this type was one of the earliest to be domesticated and was of wide
distribution in an early period of human culture. Remains of a
larger type of dog, C. intermedius, are also wide-spread in late Neo-
lithic or Bronze culture layers of middle Europe, and correspond
broadly to the larger type of Indian dog, a parallelism that is sug-
gestive of the common origin of the large and the small types of dogs
in Europe and America, probably from Asiatic prototypes.

ALLEN: DOG^S OP^ THE AMEKICAN ABORIGINES.

489

Cranial Measureinonts

.si

-y;

0 ^

GT
C

E si

to
00

o>
a t>-

B ^

'i'ji

0 P:

tjC _

c .

o .
^ Z

r X

>■ a;

Alveolus of i' to occipital condyle

—

132

140

138

142

137

139

145

" " " median edge of

palate

—

74.5

—

Alveolus of i' to orbit

—

" " " alveolus of molar 2

77.5

—

" c " " « m'^ . . .

64.5

—

" p' " " " m^ . . .

51.5

—

" ;/- " " " ?»2 ...

42.5

—

" ml " " " 7)1'- . . .

—

Length of upper carnassial, ;;■'....

16.6

14.5

14.3

15.5

16.3

Median length of nasals

—

Width across occipital condj^les . .

—

• —

" " palate at //i'

52.5

— ■

" '■ supraorbital processes

—

" " zygomata

.. .

78 '

Lower jaw, alveoli n to in^

—

" c to m,

—

74.5

—

" pitoms .

—

— -

—

" pstoms

—

48.5

—

" - " " p.tom.........

—

" 7m torn,

—

■ —

—

Length of nu

18.5

17.5

—

Skolplal Measurements

Femur

—

128

Tibia

130

Early Accounts. — Hernandez disposes of tlie Techichi in few words,
as being the third sort of dog he knew to be found in Mexico. It
must have become scarce by his time (about 1578) as he had not seen
it himseU" but describes it thus: — "Catulis simiHs est nostratibus,
Indis eduhs, tristi aspectu, ac caetera vulgaribus simihs" (similar to
our spaniels, eaten l)y the Indians, of melanchoh- visage, but other-
wise like the common dogs). J. Jonstonus, writing in 1657, includes
in his account of dogs, a transcription of Hernandez's passage as to

490 lULLETIX: Ml'SEX'M OF COMPARATIVE ZOOLOGY.

the three sorts of dogs in jNIexico. He adds further that the Indians
of Cozumel Island ate these dogs as the Spaniards do rabbits. Those
intended for this purpose were castrated in order to fatten them.

Clavigero, the historian of early Mexico, wrote that the breed was
extinct in his time, due, as he supposes, to the Spaniards' having pro-
vided their markets with them in lieu of sheep and cattle.

Possibly this breed of dog is the one mentioned in De Soto's relation
of his march through Florida. At one place the cacique of the village
sent him a present including "many conies and partridges. . . .many
dogs .... which were as much esteemed as though they had been fat
sheep." x\t another place, "the Christians being seen to go after
dogs, for their flesh, which the Indians do not eat, they ga\'e them
three hundred of these animals." Again, at a small Indian village
called Etocali, the expedition got " maize, beans, and little dogs, which
were no small relief to the people."

As late as 1805, Barton (1805, p. 12) who had made special inquiry
of William Bartram, as to the dogs of the Florida Indians, quotes
him, that the latter had in addition to the larger dogs, a smaller breed,
about the size of a fox, which probably was of the type under discus-
sion.

It is probably this dog, if not also the short-nosed Aariety, that
figures largely in the mythology of the Mayas of Yucatan. Among^
several representations of the dog in the Mayan codices are seen short-
nosed and long-nosed heads, but whether these really indicate differ-
ent breeds of dogs or different artists that made them cannot be
determined. All are shown with erect, sometimes with cropped ears,
a tail that is of medium length, usually shaggy, and recurved. Black
patches are commonly represented on the body, and the eye of the
dog often centers in a black area. Seler (1890) speaks of its use as a
sacrificial animal in Yucatan, sometimes in place of a human being.
Placed in the grave, the dog carried its master's soul across the " Chi-
cunauhapan" or nine-fold flowing stream. According to Sahagun,
some were black and white, others dark red, and there were short-
haired and long-haired dogs, but he does not state whether the small
and the large types of dogs each had short-haired and long-haired
^•arieties. A brief summary of the significance of the dog in the
religious life of the ]\Iayas is given by Tozzer and Allen (1910, p. 359).

allen: dogs of the american aborigines. 491

Hare-Indian Dog.
Plate 1, fig. 2.

i 1829. Canis lagopus Richardson, Fauna Boreali-Amer., 1, p. 78, pi. 5 (not
Canis lagopus Linne, 1758, q. e. Alopex).
1867. Canis domesticus, lagopus Fitzinger, Sitzb K. akad. wiss. Wien, 56, pt. 1,

p. 407.
Canis faviiliar is orthotus lagopus Reichenhach, Regn. anim., pt. 1, p. 13.

Characters. — A small, slender dog, with erect ears and bushy tail,
feet broad and well-haired. Color white with dark patches.

Distribution. — Formerly found among the Hare Indians and other
tribes that frequented the borders of Great Bear Lake and the banks
of the Mackenzie River.

Description. — This seems to have been a small dog, of the Techichi
type. Richardson, who gave a figure and description of it from first-
hand acquaintance, characterizes it as slightly larger than a fox but
smaller than a coyote, and apparently of rather slender proportions.
The head was .small with sharp muzzle, erect thickish ears, somewhat
oblique eyes; the tail bushy and sometimes carried curled forward
over the right hip, though this does not appear in Richardson's figure;
foot broad and well-haired. He describes an individual as having the
face, muzzle, belly, and legs white; a dark patch over the eye, and on
the back and sides, larger patches of dark blackish gray or lead color,
mixed with fawn and white. Ears white in front, the backs yellowish
gray or fawn; tail white beneath and at the tip.

Notes. — It seems probable that this small breed was lost in the
early part of the last century. At all events, writers subsequent to
Richardson do not seem to have met with it, and those that mention
it, seem to have confused it with the Common Indian Dog. Thus
B. R. Ross (1861) and Macfarlane (1905, p. 700) clearly had in mind
a different animal; and a skull sent by the latter to the U. S. N. M.
as lagopus (from Fort Simpson, Mackenzie River) is a large dog,
evidently the Common or Larger Indian Dog. Hamilton Smitli
(1840, p. 131) takes his description in part from Richardson, and
mentions a pair of these dogs as then living in the Zoological Society's
Gardens at London. Audubon and Bachman likewise are indebted
to Richardson for their account, though their figure, by J. W. Audu-
bon, is said to be from a stuffed specimen, perhaps one of those previ -

492 bulletin: museum of comparative zoology.

ously li\ing in the Zoological Society's Gardens. The dimensions
they gi\e however, seem rather large.

Richardson says further that it was used solely in the chase and
was probably too small to serve as a burden carrier. Its voice was a
wolf-like howl, but at some unusual sight it would make a singular
attempt at barking, commencing AVith a peculiar growl and ending
in a prolonged howl.

Here may be mentioned what seems to be an unknown or vanished
breed of dogs as indicated in the account of Frobisher's ^'oyage to
Arctic America in 1577. At the present Frobisher Bay, in south-
eastern Baffin Land, the expedition found in addition to the large
dogs used for sledging, a smaller breed, which was apparentl\' used
only as food, and allowed the freedom of the skin tents of the Eskimos.
The historian of the expedition writes that they "found since by
experience, that the lesser sort of dogges they feede fatte, and keepe
them as domesticall cattell in their tents for their eating, and the
greater sort serve for the use of drawing their sleds." At York Sound,
the same writer relates that on going ashore to examine "certaine
tents of the countrey people," they "found the people departed, as it
should seeme, for feare of their comming. But amongst sundry strange
things which in these tents they found, there was rawe and new killed
flesh of unknowen sorts, with dead carcasses and bones of dogs"
(Hakluyt's Voyages, Everyman's Library, ed. 5, p. 212, 215). Concern-
ing this "lesser sort of dogges," nothing further seems to be known,
whether they were a dwarf variety of the Eskimo dog, or as seems
likelv, a small breed similar to those of the Hare Indians or of other
tribes of the mainland.

FuEGiAN Dog.
Plate 4, fig. 2.

Characters. — Size small, as large as a terrier, muzzle slender, ears
large, delicate, and erect, body and limbs well-proportioned, shoulders
higher than rump; tail long, drooping, slightly recurved at the tip
and well-fringed; feet webbed; color uniform grayish tan, or often
with patches of black or tan, and areas of white; inside of the mouth
dark-pigmented.

Distribution. — Found chiefly among the "Canoe Indians" — Yah-
gans and Alacalufs — of the Fuegian Archipelago, from Cape Horn to
Beagle Channel, and northwestward, probably at least to the western
part of Magellan Strait.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 49^^

Descriptions. — The best account of the Fuegian Dog is that given
l)y d'Herculais (1884) of two Yahgan Dogs brought back to France
by Dr. Hyades of the Mission scientifique au Cap Horn (expedition
de la Romanche), in 1883. These were obtained as puppies from
the Yahgans at Orange Bay and grew up to be tame and affectionate
dogs. They are described as small but well-proportioned, remarkable
for their large pointed and erect ears, and very sharp slender muzzles.
The color-pattern is very variable, often a uniform grayish tan recall-
ing the jackal; again, the body is marbled with extensive black or tan
areas on a white ground. The feet are plainly webbed. The two
dogs above referred to, were said to measure, the male and female
respectively: — height at shoulder, 49 and 44 cm.; length from tip of
nose to root of tail, 80 and 72 cm. ; length of tail, 2G and 23 cm.

External Measurements. — Dechambre (1891) in a note on these
same dogs, gives the following dimensions, evidently of a female: —

Scapuloischial length 52 cm.

Height at shoulder 41 "

Height at rump 39 "

Height at axilla 25 "

Thoracic perimeter 58 "

Distance between ears 9 "

" " inner corners of eyes 4.5 "

outer " " " 8.5 "

Breadth of forehead 11 "

Length of head 22 "

" " muzzle 9 "

Interorbital width at outer corner of eye 9.5 "

The further description by Dechambre supplements that of d'Her-
culais based on the same individual. He describes its fox-like head
with pointed muzzle, broad forehead, its erect and high-set ears,
usually directed forward, very mobile; eyes slightly oblique. The
body is large, limbs slender, the neck short and powerful, the
shoulders slightly higher than the rump; tail bushy and carried
high. Pelage with a short imder fur, pied black and white, passing
to slaty at the throat, clouded with tan ; over each eyebrow a white
spot with a few fulvous hairs. The coat has the appearance of a
domesticated animal in its pattern.

Captain Fitzroy of the Beagle, in a letter to Hamilton Smith (1840,
p. 214) describes these dogs of the 'Canoe Indians' as resembling
" terriers, or rather a mixture of fox, shepherd's dog, and terrier. All

494 bulletin: museum of comparative zoolooy.

that I examined had black roofs to their mouths, but there was much
variety in the colours and degrees of coarseness of their coats. * * *
Many Fuegian dogs are spotted and not a few have fine short hair,
but all resemble a fox about the head. * * * One brought from Tierra
del Fuego Mas white with one black spot, and very handsome; his size
was about that of a terrier, his coat short but fine, and his ears extremely
delicate and long, although erect;" the muzzle also is long, the tail
rough and drooping.

Skull and Limh-botics. — In a recent paper. Professor Lonnberg
(1919) has given what appear to be the first published hgures and
measurements of the limb-bones and skull of this dog. His speci-
men was a skeleton obtained by Xordenskjold in 1895-96 during his
Tierra del Fuego expedition. As this author demonstrates, the skull
is that of a true dog, and shoAVS no relationship with the native canid,
Pscitdaloprx lycoidcs. A comparison of the cranial measurements
with those given for the Techichi of North and South America, shows
a very close approximation, amounting almost to identity. The
first lower molar in the Fuegian Dog seems smaller, however, 16.5
mm. in Lonnberg's specimen against 17.5 to IS. 5 mm. in the more
northern dogs. For better comparison, the following measurements
of the Fuegian Dog are reproduced from this paper (Lonnberg, 1919,
p. 11):-

Condylo-incisive length 141 mm.

Length of palate 71 . ;3 "

Front of canine to back of m- 64 "

Length of premolar* 15. 2 "

Length of upper premolar-molar series. ... .51 "

Width of palate outside m^ 52 . 6 "

Zygomatic width 81 "

Length of nasals mesially 46 "

Length of lower mi 16. 5 "

Length of humerus 105 "

Length of ulna 125

Length of femur 132

Length of tibia 139

l.scs. — The Fuegian Dog is acti\ e and strong in proportion to its
small size; quiet, faithful to its master, and able to withstand much
privation; A-igilant and extremely sly. It is capal)le of barking like
the European dogs.

They are of invaluable service to their masters in hunting, particu-
larly in the pursuit of otters (Lutra fclina), which are assiduously

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 495

sought. Indeed Fitzroy wrote that " it is well ascertained that the
oldest women of the tribe are sacrificed to the cannibal appetites of
their countrymen rather than destroy a single dog. ' Dogs,' say they
'catch otters; old women are good for nothing.'" They are vigilant
watch-dogs, liarking furiously at a stranger. Their small size, and
consequent adaptability as canoe companions, are no doubt the chief
cause for their preference by the Canoe Indians of the west Patagonian
Archipelago, over the larger dogs foinid among the so-called Foot
Indians of the mainland and the eastern and inland parts of Tierra
del Fiiego.

Remarks. — In the absence of specimens for comparison, it is not
altogether clear that the Fuegian Dog can be satisfactorily distin-
guished except in minor particulars from the Techichi or Alco of Peru
and Alexico. ^Molina apparently thought it identical. In general it
appears closely similar, but perhaps of more slender build, a bushier
tail with recurved tip, well-palmated feet and a shaggier coat, though
Fitzroy speaks of variation in this last character.

In his Bibliography of the Fuegian tribes, Cooper (1917, p. 186)
has summarized the references to dogs in the literature referring to
these people. As early as 1557, or perhaps 1553, the Chonos at the
northern end of the Chilian Archipelago, were credited with having
dogs, as appears from Goicueta on the authority of Cortes Hojea.
The first mention of dogs in the Strait of Magellan appears^ to be
that of Narbrough, who in 1670, found the natives of the Elizabeth
Islands in possession of large mongrel dogs of several colors. He
compared them to the race of Spanish dogs he had fovmd among the
Patagonians of Port Julian. Probably these were not of native stock.
Twenty-six years later de Gennes saw five or six small dogs among the
Alacalufs of Port Famine. The Manekenkn met by the first Cook
expedition in 1 769 at Good Success Bay, southeast end of Tierra del
Fuego, had dogs about two feet high with sharp ears; they all barked.
The small dog here described is apparently found among the so-called
Canoe Indians of the western archipelago, the Yahgans and Alacalufs,
the most southerly tri])es of men in the world.

SUOKT-NOSED I.\DL\N DOG.

Plates 6, 11.

1885. Pachycyon robustm J. A. Allen, Mem. M. C. Z., 10, 13 pp., 'S pis.
1885. Canis ingae veriagus Nehring, Sitzb. Gesellsch. naturf. freunde B(U-liii,
p. 5-13 {not Canis familiaris vertagtisljinnc , Syst. nat., 12th ed., 1766, 1, p.
57.

496 bulletin: museum of comparative zoology.

Characters. — A stoutly built dog, the size of a small terrier, witii
erect ears, short heavy muzzle, high forehead, short body and limbs,
well-developed tail.

The color seems to have been black and white; sometimes more
uniformly black, or yellowish with dark blotches.

The skeleton is stoutly proportioned, the limb-bones short and
thick, the humerus with a very small or no olecranal perforation.
The sagittal crest is chiefly developed at the occiput. Correlated
with the slight reduction of the maxillary bones, and the widening
of the palate, is the fact that the last molar is placed just in advance
of a transverse line through the posterior boundary of the palate.

Distrihution. — Skeletal remains of this peculiar small dog ha\e
been found in ^'irginia in a superficial cave-deposit, as well as in the
shell-mounds of San Nicolas Island on the coast of southern California.
A well-preserved dried or mummified example was lately discovered
by Mr. S. J. Guernsey in a l)urial antedating the Cliff Dwellers, in
the Marsh Pass region of Arizona; and Reiss and Stiibel have dis-
covered its mummified remains in the prehistoric necropolis of
Ancon, Peru (see Nehring, 1884b). In the M. C. Z. is a humerus
lacking the epiphyses, of a young specimen from Pecos, New Mexico,
obtained by Dr. A. V. Kidder. These localities may be taken as
limiting the known extent of its distribution.

Notes. — In 1885, Dr. J. A. Allen described as a new genus and
species Pachycyon rohustus, an extinct type of dog from Ely Cave,
Lee County, Virginia, basing his account upon a pelvis, a femur, a
tibia, a scapula, and a humerus of which he publishes excellent illustra-
tions. These bones were obtained in the course of excavating the
superficial layer of earth on the cave-floor, and though it is not certain
exactly at what point they were found, no excavations deeper than a
foot were made. Remains of Indian occupation were numerous, and
other bones were obtained in the cave. There is nothing to indicate
great age in the type-specimens fM. C. Z. 7,091); indeed the bones
are quite fresh in appearance, only slightly discolored with earth.
They are chiefly notable for their small size and rather heavy ungrace-
ful proportions, while the humerus is particularly marked on account
of its lacking the usual perforation over the middle of the epicondyle.
This perforation is almost always present in Eurasian dogs, as well as
in coyotes and wolves. No further light has since been shed on the
nature of this animal nor have any parts of its skull been found.

Among the remarkable discoveries made by Mr. S. J. Guernsey in
the course of archaeological exploration in the Marsh Pass region of

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 497

Arizona for the Peabody Museum, were the dessicated remains of
two dogs with human burials of an age apparently antedating the
culture of the Cliff Dwellers. One of these dogs is small, about the
size of a Fox-terrier but more compactly and heaxily built, with a
shorter head, erect ears, and longer tail. It still shows a black and
white pattern, with a narrow median white line from nose to fore-
head, a white chin, throat, and lielly, a white collar, white feet, and
tail tip. Much of the body is black. In the length of the limb-bones
and pelvis as nearly as can be determined from careful study of the
dried and mummified specimen, it corresponds exactly with Pachy-
cyon. By making incisions through the dried tissue at the elbow, it
was possible to lay bare the olecranal cavity above the joint where
the large perforation is usually present. It was found that in the
right humerus a small perforation was present, about 3 mm. in diam-
eter, while in the left humerus there were merely two small pores side
by side. The animal was young, still retaining a milk incisor, and so
it is likely that had it been as old an individual as the one whence the
type-bones of Pachycyon were derived, these foramina would have
ossified completely, perhaps leaving, as in the type-humerus, a shallow
pit in the posterior side of the olecranal fossa, as an indPcation of the
foniier perforation. So complete is the correspondence of the bones
of Pachycyon with those of this prehistoric dog of Arizona that they
may be unhesitatingly pronounced those of a similar if not identical
breed of Indian dog.

Not less interesting is a comparison of the humerus of Pachycyon
with a humerus figured by Nehring (18S4b, Plate 118, fig. 4, 4a) from
a mummified dog exhumed with human-mmnmies in the ancient
necropolis of Ancon, Peru. In measurements, there is practical
identity as shown in the following table (the measurements of the
Ancon humerus are taken directly from Nehring's figure, of natural
size) : —

Pachycyon Ancon

Greatest length of humerus 97 mm. 97 mm.

Greatest diameter through head of humerus 31 . 5 29 . 5

Transverse " " " " " 21 24

Transverse diameter 6i distal end of same 25 25

Nehring's figure shows substantially the same type of thick stout
humerus, and as he remarks, has the further peculiarity of lacking
any trace of perforation of the olecranon fossa. It should be added
that the humerus, shown in his figiu-e is nevertheless \ery slightly

498 bulletin: museum of comparative zoology.

more bowed than that of the type of Pachycyon, and in his opinion
the Peruvian Dog corresponded closely to a European Turnspit or
Dachshund, whence he calls it Canis ingae vertagus. The figures of
the skull of the same specimen likewise show an apparent similarity
in outline and proportions to that of the Arizona mummy.

There seems thus to be no doubt that Pachi/ci/on robustiis is after
all only a breed of dog cultivated by the Indians of the southern
parts of North America and of Peru. It is therefore no longer to be
thought of as a problematical mammal of the Pleistocene.

Among the dog-bones obtained by the University of California's
in^■estigations of the Indian shell-moimds on San Nicolas Island, off
the coast of southern California, are two crania nearly identical in
measurements with the Marsh Pass specimen that appear to repre-
sent this same small, short-nosed dog. They are characterized by
their broad brain-cases, spreading zygomata, wide palates, shortened
rostra, and small teeth. In profile the dorsal outline of the brain-
case is gently rounded, not fiat. The shortness of the rostrum does
not amount to real deformity however, for the lower jaw closes nor-
mally into its place and the premolars are not markedly crowded,
though p^ is turned at an angle of nearly 50° from the axis of the skull
to adapt its position to the sudden narrowing of the skull at this point.
Premolars 1 and 2 are normal in position, and there is a short diastema
between p^ and the canine. The ossification seems particularly
heavy, yet though old, neither skull has de\"eloped a sagittal crest
except at the interparietal region. In the dried mummy from Marsh
Pass, the shortened nose and elevated forehead give a characteristic
appearance to the head which is evident in these crania as well. No
liml)-bones that can be assigned to this dog, ha\"e appeared among the
Calif ornian collections. In both crania the opening of the posterior
nares is narrow, and a transverse line drawn at right angles to the
cranial axis at the posterior end of the palate falls behind the last
molar, iufiicating deviation from the normal condition.

The following skull-measurements show close agreement. One of
the Calif ornian crania (r^is^) lacks any trace of the alveoli of ?»-
which are partly broken and partly resorbed. The first premolar is
wanting also. The proportions of the maxilla are, however, practi-
cally the same in both specimens. The Ancon specimen is figured
by Nehring (lS84b) of natural size and the measurements are taken
from this figure. It too lacks the first upper premolar, and in every
respect confoniis to the appearance of the other crania.

ALLEN: DOGS OF THE AMERICAN ABORIGINES.

499

Measurements of the Skull

Peru:
A noon

Ariz.:

Marsh

Pass

16,355
Calif.

141

?132

138

123

—

— •

—

71.5

—

16.5

—

• — ■

—

16,356
Calif.

Greatest length, occiput to median incisor

(alveolus)

Greatest length, edge of foramen magnum to

median incisor

Median incisor to edge of palate

" " " orbit (anterior edge)

" tn^ (alveolus)

Canine " m^ "

Premolars '"■' (alveoli)

Length of premolar ^

Molars '~- (alveoli)

Width of palate outside m.^

a n (1 u ,_:i

Zygomatic width

Mastoid width

Width of occipital condyles

Nasals, length

138

121

68
54

56.5

In addition to the limb-measurements given on p. 497, the Arizona
mummy gives the following: — total length from tip of nose to tip of
tail following curve of back, 705 (circa); tail about 195; ulna 120
(circa); carpus to end of longest claw 90; ear about 60-70 mm. long
including hair; tail 195; femur 106 (circa); tibia 116 (circa); hind
foot 122.

Rrmarks. — Although this type of dog seems to have been wide-
spread among the aborigines of southern North America and north-
eastern South America, it appears to have quite disappeared and is
not clearly identifiable in any of the accounts of the early writers.
Mr. Guernsey's discovery of a well-preserved mummy in a burial of
considerable age in Arizona, has confirmed my previous identification
of the Virginia bones of Pachycyon w'ith those of Nehring's short-
limbed dog-mummy of Ancon. The cranium is characterized by its
breadth and stoutness, its shortened snout and high forehead, gently
convex dorsal profile of the brain-case, and the small teeth (upper
carnassial 16-17 mm.). The Calif ornian crania agree substantially
in every detail. Probably this is the same dog that Moore (1907,
p. 423) disco\ered in Indian mounds on Crystal River, west Florida,
of which Lucas observed, " the front of cranium of carnivore and jaws,

500 bulletin: museum of comparative zoology.

are from the same animal, the short-faced dog something hke a bull-
terrier that seems to have been a favorite with the Indians of the
southwest".

Peruvian Pug-nosed Dog.

Plate 12.

1885. Canis ingae molossoides Nehring, Sitzb. Gesellsch. naturf. freunde
Berlin, p. 5-13.

Characters. — Similar to the Short-nosed Indian Dog but with even
shorter facial bones, an undershot lower jaw, broader zygomata and
posterior narial passage. The increased shortening of the face causes
a slightly more elevated forehead. The color seems to have been
yellowish or whitish, marked or clouded with dark brown.

Distribution. — This Dog is known only from the Peruvian High-
lands, where its remains have be*n found with ancient burials of the
aborigines at Ancon and Pachacamac.

Skull-Characters. — A comparison of six skulls from Peru (loaned by
the U. S. N. M.) with those of the Short-nosed Dog of North America,
leaves little doubt that the Peruvian Pug-nosed Dog is derived from
the latter, perhaps through some sort of cross-breeding, possibly as
an occasional result of a particular cross, or through the dominance
of its peculiarities in cross-bred animals. In most respects, the skulls
of both are essentially alike, but the shortening of the rostral portion
in the present breed is more pronounced, resulting in an undershot
lower jaw. Yet the reduction of the maxillaries is not so extreme as to
cause very great crowding of the premolars as in our Bull-dogs or the
Pekinese Lap-dogs. Thus in two out of six crania, the third premolar
is set almost transversely to the long axis of the skull, but in the
others it retains about the usual relation. The second premolar, in
two cases, is turned inward at more than the usual angle. In only
one of the six skulls is the first upper premolar missing, and here on the
left side onlj'.

The opening of the posterior nares is, very wide in comparison with
the common Short-nosed Dog, and the zygomatic arches are broader.
In none of the six skulls do the temporal ridges unite to form a median
crest except at the occiput along the interparietal bone. On account
of the shortening of the facial bones, the forehead is high, with a deep
and broad groove medially. A further result of this shortening is
the greater upward turn of the palate, best seen when the crania are

ALLEN: DOGS OF THE AMERICAN ABORIGINES.

501

on a flat surface. The palate of the Pug-nosed Dog, makes an angle
with the table of about 27° against about 15° in the case of the longer-
nosed breed. The same rugose surface of the brain-case, the lieaviness
of bone and the thickened prominc/twes at each side of the posterior
narial openings, characteristic of the Inca Dog, are seen in this breed
as well.

No Hmb-bones have been obtained that can be referred to this dog,
but it is Hkelv that thev were short and thick hke those of the related
breed.

The following table gives dimensions of the sL\ skulls in the U. S.
N. M. and is interesting for comparison with those of the Short-nosed
Indian Dog.

Measurements of the Skulls

U.S.

N. M.

172,885

172,883

172,886

172,887

172,884

176,307

Occipitorostral length (excluding
incisors)

124
104

16.5

15.5

102

138
121

15.5

14.5

15.5

109
30

138
125

16.5

17.5

15.5

94
29

142
119

67.5

17.5

145

Basal length

125

Palatal length

Orbit to tip of premaxillary

Upper tooth-row

(alveoli)

Front of canine to back of molar -

(crowns)

Front of canine to back of molar ^

(alveoli)

69
56.5

Length of premolar^ (crown) ....

" . " " (alveolus) ..

" " molars^"^ (crowns) ....

" " (alveoli)

Lower tooth-row (alveoli)

Zygomatic width

16.5
15
17

16.5
81
102

Breadth of occipital condyles ....

31.5

Remarks. — ^The existence of this breed of aboriginal dogs with
shortened face and undershot, bull-dog-like jaw, was first discovered
by Reiss and Stiibel in the course of their investigation of the necro-
polis of Ancon, Peru. Nehring (1885) published an account of their
discovery and gave the Latin name Canis ingae molossoides to the

502 bulletin: musei'm of comparative zoology.

l)reed. At first but a single specimen was found among numerous
other dog remains, but further search brought a few more to Hght,
and more recently the Yale-National Geographic Society P^xpedition
has recovered several skulls, from Huacho and Pachacamac.

The presence of this pug-nosed doj* among the ancient Peruvians
is doubly interesting, not only in that this \ariation should have
occurred here, apparently cjuite independent of similar cases in the
Old World, but in that it should have been preserved, whether through
accident, or as supposed, through purposeful selection. Such a
shortening of the face through the imperfect development of the bones
of the rostrum is found occasionally in other domesticated manunals.
The short-faced Cheshire Hogs and similar l)reeds fiu'nisli like in-
stances of the selection and preservation of this mutation, which
appears to be definitely heritable. Among imdomesticated species,
the case of a European Fox is recorded by Donitz (1(S69) in which the
rostrum was shortened abnormally, producing a bull-dog-like appear-
ance, with undershot jaw. The second and third premolars of the
upper jaw were opposite the third and fourth respectively of the lower
jaw, while the upper canine fitted into a space between the first and
second lower premolars. Schmitt (1903) agrees with Studer (1901)
that such cases are due to the retention of embryonic conditions but
considers them to be a result of domestication. This, however, is
not necessarily the case, as the above instance shows. The case of a
"bull-dog-headed calf" is recorded by Warren (1910) as having ap-
peared as a " sport " \'ariation.

Notwithstanding the comparatively high cultural development of
the Incas, it nuiy be doubted whether they purposely bred these dogs
for their peculiarity of face. Quite as likely the anomaly arose,
perhaps as a frequent result of cross-breeding between certain of the
other canine races, or as a local abnormality, which as a Mendelian
character, frequently cropped out in chance crosses. This may be
indicated by the apparent rarity of this type of dog in the Ancon
burials, and by the considerable variation in slight details of the form
of the skull, as if no special type were bred for.

An interesting anomaly of an opposite nature is worth recording in
this connection, namely that of a Jackal shot by Dr. J. C. Phillips in
Arabia (M. C. Z. 15,872) in which the wider jaw has failed to reach
its normal length and is overshot by the upper jaw. The lower canine
closes hchind the upper instead of anterior to it as in normal cases.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 503

Summary.

Recent careful studies of the teeth indicate that the domestic dog's
relationship is with the wolves rather than with the groups of canids
represented l)y coyote, jackal, or fox. The ultimate wolf-like ancestor
of the dog is yet to be determined, but present evidence favors the
view that it was not one of the large circiunboreal wolves, but possibly
a distinct and smaller species, from which both large and small breeds
of dogs have been derived. -

The domestic dogs of the American aborigines were ciuite as trulj'
typical dogs as those of Asia, and may be assumed to ha\e reached
America from that continent, with their human companions. Al-
though it is possible that the larger dogs may interbreed occasionally
with wolf or coyote, there is no good reason to suppose that such cross-
ing has had much if any, influence on the original stock.

■ In a very general way, three types of dogs nia\' be distinguished
among the American aborigines: (1) the large, l)road-muzzled, Es-
kimo Dog, with hea\y coat and tail curled forward over the hip;
(2) a larger and (3) a smaller Indian Dog, from which are probably to
be derived several distinct local breeds. Of the larger style of dog as
many as ele\en \arieties may perhaps be distinguished: of the smaller,
five.

An interesting and suggesti\'e parallel is found among prehistoric
European dogs, of which in late Neolithic and early Bronze periods
there were a large and a small type — Canis intermedium and C.
palustris — corresponding rather closely to the Larger or Common
Indian Dog and the Small Indian Dog or Techichi. The obvious
probability is that these two general types of dogs were then widely
cultivated in Asia, and at a very early period reached Europe and
America with the human immigrants. In a similar way the Eskimo
Dog is of a type common to northern Asia and Europe, and doubtless
reached America with the Eskimos, whose arrival, at least in eastern
America is usually regarded as relatively recent.

504 BT'LLETIN: MITSET'M OF COMPARATIVE ZOOLOGY.

BIBLIOGRAPHY.

Acconci, L.

1880. Continuazione dello stutlio dei resti fossili rinvenuti nella oaverna
di Cucigliana. Ordine dei carnivori: famiglia dei Canidi. Atti Soc.
Tosc. sci. nat. Pisa, Proc. verb., 2, p. 41-42.

Acosta, J. de.

1590. Historia natimil y moral de las Indias, etc. Sevilla, 8vo, 536 pp.
and index.
Allen, J. A.

1885. On an extinct type of dog from Ely Cave, Lee County, Virginia.
[With a note] on the age of the Ely Cave by N. S. Shaler. Mem.
M. C. Z., 10, 13 pp., 3 pis.
Anderson, R. J.

1901. Abnormal dentition in the dog. Irish nat., 10, p. 89-90.
Anutschin, D. N.

1881. [Ueber die Hunderassen der Steinperiode an den Ufern des Lado-
gasees]. 5th Archaol. congr. Tiflis. (Russian).

1882. [Zwei Rassen des Hundes aus dem Torfmooren des Ladogasees].
Moskou. (Russian). See Arch, anthrop., 1882, 14, p. 333.

Audubon, J. J., and Bachman, J.

1849-54. The quadrupeds of North America. Xew York, 8vo, 3 vols.,
illus.
[Bancroft, Edward].

1769. An essay on the natural history of Guiana. London.
Bannister, H. M.

1869. The Esquimaux dog. Amer. nat., 3, p. 522-530.
Bartlett, A. D.

1890. Observations on wolves, jackals, dogs, and foxes. Proc. Zool.
soc. London, p. 46-48.
Barton, B. S.

1805. Some account of the different species and varieties of native Ameri-
can, or Indian dogs. Barton's Med. andphy,s. journ., 1, pt. 2, p. 3-31.
Bartram, William.

1792. Travels through North and South Carolina, Georgia, east and
west Florida, etc. London, 8vo., illus.
Baudouin, M.

1910. Lc chien en prehistoire. Bull. Soc. prchist. de France, 7, p. 314-
316.
Blainville, H. M. H. Ducrotay de.

1841. Osteographie ou description iconographique comparee du squelette
et du systeme dentaire des cinq classes d'animaux vertebres recents et
fossiles pour servir de base a la zoologie et a la geologie. Mammiferes,
Carnassiers, vol. 2. Paris, 4to, illus.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 505

Blanford, W. T.

1877. [Professor Jeitteles' researches on domestic dogs]. Proc. Asiatic
soc. Bengal, p. 114-117.

Boule, Marcellin.

1889. Les prcdocesscurs de nos canides. Compt. rend. Acad. sci. Paris,
108, 1). 201-203.
Bourguignat, J, B.

1875. Recherches sur les ossements de Canidae constates en France a
Tc'tat fossile pendant la poriode qiiarternaire. Ann. sci. geoL, 6, art.
6, 60 pp., pi. 16-18.
Brown, Robert.

1868. On the mammalian fauna of Greenland. Proc. Zool. soc. London,

p. 330-362.
1875. The same, reprinted, with cori'ections and annotations by the
author. 34 pp.
Browne, Patrick.

1789. The civil and natural history of Jamaica. London, 4to.
Biirton, L. A.

1917. Raising wolf dogs. Forest and stream, 87, p. 448.
Catlin, George.

1841. Letters and notes on the manners, customs, and condition of the
North American Indians. 8vo., 2 vols., illus.
Children, J. G.

1827. On the Esquimaux dog. Zool. journ., 3, p. .54-56, pi. 1.
Colenso, William.

1878. Notes, chiefly historical, on the ancient dog of the New Zealanders.
Trans, and proc. New Zealand inst., 10, p. 135-155.

Cooper, J. M.

1917. Analytical and critical bibliography of the tribes of Tierra del
P\iego and adjacent territory. Bull. 63, U. S. bur. Amer. ethnol.
Cope, E. D.

1879. On the genera of Fehdae and Canidae. Proc. Acad. nat. sci.
Phila., p. 168-194.

1879a. The Japanese Lap Dog {Dysodus -pravas). Amer. nat., 13, p.
655-656.

1 883. On the extinct dogs of North America. Amer. nat. , 17, p. 235-249,
fig. 1-14.

1887. The pug-dog and the Chihuahua dog. Amer. nat., 21, p. 1125-

1126.
1893. [On a jaw of an Indian dog] in C. B. Moore's Certain shell heaps of

the St. John's River, Florida, hitherto unexplored. Amer. nat., 27,

p. 614, 3 figs.
Cope, E. D., and Wortman, J. L.

1884. An account of the mammalian fauna of the PostpHocene deposits,
in the state of Indiana. 14th ann. rept. state geol. Indiana, pt. 2, p. 4-62,
pi. 1-6. (See p. 8, footnote).

506 bulletin: museum of comparative zoology.

Coues, Elliott.

1873. The prairie wolf , or coyote : Caiiis latrans. Amer. nat., 7, p. 385-

389.
1876. Reversion of the dog to the feral state. Forest and stream, 7,
p. 213.
Cunningham, R. O.

1S71. Notes on the natural history of the Strait of Magellan and west
coast of Patagonia made during the voyage of H. M. S. ' Nassau ' in
the years 1866, 67. 68, 69. Edinburgh, 8vo., illus.
Cuvier, F.

1808. Observations sur le chien des habitans de la Nouvelle-HoUande,
prccedees de quelques refle.xions sur les facultes morales des animaux.
Ann. Mus. d'hist. nat. Paris, 11, p. 458-476.

1810. Recherches sur les differences d'organisation qui existent entre les
races des chiens domestiques. Xouv. bull, des sci. Soc. philom. Paris.
2, p. 185-187.

1811. Recherches sur les caracteres osteologiques qui distinguent les
principales races du chien domestique. Ann. Mus. d'hist. nat. Paris.
18, p. 333-353, i>l. 18-20.

Dechambre, Paul.

1891. Note sur les chiens de la Terre de Feu. Mission sci. du Cap

Horn, 1882-1883, 7, p. 391-392.
1894. Race canine. Classification et jjointage. ^Icm. Soc. zool.
France, 7, p. 331-362.

Dbnitz,

1869. [I^eber eine monstrose schiidelbildung eines fuchses aus Schlc-
sien]. Sitzh. Ces. naturf. freunde Berlin, 1868, p. 20-21.
Duerst, J. U.

1904. Die tierwelt der ansiedelungen am schlossberge zu burg an der
Spree. Arch, anthrop., new ser., 2, ]). 233-294, pi. 16-20.
Duges, Alfredo,

1882. E\ i)erro de Chihuahua. La naturaleza, 5, p. 14-17, fig. 1-3.
Eastman, C. R.

1916. Hunting-dogs of the Ancients. Amer. mus. journ., 16, p. 403-
408. 6 figs.
Eaton, G. F.

1898. The prehistoric fauna of Block Island [R. I.], as indicated by its

ancient shell-heaps. Amer. journ. sci.. ser. 4, 6, p. 137-159, pi. 2, 3.
1916. The collection of osteological material from Machu Picchu. Mem.
Conn. acad. arts and sci., 5, 96 pp., 39 pis., 2 tables, map.
EifEe, O. E.

1909. Kreuzung von polarhund mul Dingo. Zool. beobachter (Zool.
garten), 50, p. 312-313.
Ermel, Alexander.

1889. Eine raise nach der Robinson-Crusoe-Insel. Hamburg, 8vo,
134 pp., 11 pis., map.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 507

Filhol, Henri.

L883. Observations relatives aux chiens actuels ct aux cariiassiers fos-
siles s'eii rapproohant le plus. Arch, Mus. d'hist. nat. Lyon, 3, p. 70-97,
])1. 5.
Fischer, J. B.

1829. Synopsis Maniinaliuni. .Stuttgart, Svo.
Fitzinger, L. J.

186C. ITntersuchungen iiber die abstammung des hundes. Sitzl). K.

akad. wiss Wien, math.-nat. el., 54, pt. 1, p. 396-457.
1867. Die ragen des zahmeu liundes. Sitzb. K. akad. wiss. Wien,
math.-nat. cl., 56, i)t. 1. p. 377-507, 514-585, 776-823.
Gervais, Paul.

187."^. Remarques au sujet du ehien domestique. .lourn. zool., 4, p. 119.
Gibbs, George. See Suckley, G., and
Gidley, J. W.

1913. Prchminary report on a recently discovered Pleistocene cave
deposit near Cumberland, Maryland. Proc. V. S. X. M., 46, p. 93-
102, fig. l-8a.
Glahn, H. C.

1784. Om den Gr<jl)nlaii<lske hund. Xye samlung, K. Xorske vid.-selsk.
skr. Trondhjem, 1, p. 485-496.
Gray, J. E.

1867. On the skull of the Chinese pug-nosed spaniel or lap-dog. Proc.

Zool. soc. London, p. 40-43, fig. 1-4.
1869. Xote on the varieties of dogs. Ann. mag. nat. liist., ser. 4, 3,
p. 236-240.

Guyon, .

1866. Des animaux disjjarus de la Martinique et de la Guadeloupe depuis
notre etablissement dans ces iles. Compt. rend. .\cad. sci. Paris, 63,
p. 589-597.
Hammeran, A.

1895. Zur geschichtc dcr lnmde-ra.s.sen. Zool. garten, 36, p. 12-17,
53-59, 88-92, 110-117, 147-153.
Harlan, Richard.

1825. Fauna Americana: being a description of the mammiferous ani-
mals inhabiting Xort h America. Philadelphia, 8vo.
Harmon, D. W.

1820. A journal of voyages and travels in the interior of X^orth America,
[etc.]. Andover, 8vo, 4.32 pp.
Harting, J. E.

1884. Dogs: ancient and modern. Zoologist, ser. 3, 8, p. 393-411.
Heame, SamueL

1796. A journey from Prince of Wales's Fort, in Hudson's Bay, to the
northern ocean .... in the years 1769 . . . .-1772.
Herculais, J. K. d'.

1884. Les chiens des Fuegiens. Science et nature, 1, sem. 1, p. 137-140,
fig-

508 bulletin: museum of comparative zoology.

Hernandez (or Fernandez), Francisco.

1651. Historiae animalium et mineralium Novae Hispaniae. Liber
iinicus in sex tractatus divisus. Rome, folio, 96 pp. (in Recchi and
Lynceus, 1651).
Hilzheimer, Max.

1905. Variationen des canidengebisses mit besonderer beriicksichtigung
des haiishundes. Zeitsch. morph. u. anthrop., 9, p. 1-40, pi. 1-5.

1908. Beit rag zur kenntnis der nordafrikanischen schakale nebst
bemerkungen iiber deren verhaltnis zu den haushunden, insbesondere
nordafrikanischen iind altagyptischen hunderassen. Zoologica, 20,
no. 53, 111 pp., 4 tables, 10 pis.
Hind, H. Y.

1859. North-west territory. Reports of progress; together with a pre-
liminary and general report on the Assiniboine and Saskatchewan
exploring expedition, made under instructions from the Provincial
Secretary, Canada. Toronto, 4to, illus.
Holland, W. J.

1908. A preliminary account of the Pleistocene fauna discovered in a
cave opened at Frankstown, Pennsylvania, in April and May, 1907.
Ann. Carnegie mus., 4, p. 228-2-33, pis. 57, 58.

HoIIard, .

1859. Nouvel exemple de croisement fecond du Cards lupus et du Cants
fumiliaris. Compt. rend. Acad. sci. Paris, 48, p. 1072-1073.
Hue, E.

1906. [On Canis le mirei, a neolithic dog from Jura]. Bull. Soc. prehist.
de France. (Not seen).

Hunter, John.

1787. Observations tending to shew that the wolf, jackal, and dog, are

all of the same species. Phil, trans. Roy. soc. London, 77, p. 253-266.

1789. A supplementary letter on the identity of the species of the dog,

wolf, and jackal. Phil, trans. Roy. soc. London, 79, pt. 1, p. 160-161.

Hutton, F. W.

1898. Note on the ancient Maori dog. Trans, and proc. New Zealand
inst., 30, p. 151-155, pi. 15.
Huxley, T. H.

1880. On the cranial and dental characters of the Canidae. Proc.
Zool. soc. London, p. 238-288.
Hyades, P. and Deniker, Joseph.

1891. Mission scientifique du Cap Horn. 1882-1883. 7. Anthropologie,
Ethnographic. Paris, 4to, illus.
Ihering, Hermann von.

1913. Le chien domestique des Calchaquis. Revista Mus. de La Plata,
20, p. 101-106, 3 figs.
Jeit teles, L. H.

1872. Die vorgeschichtlichen alterthiimer der stadt Olmiitz und ihrer

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 509

Umgebuiig. Mitth. Anthrop. ges. Wien, 2, p. 18-32, 53-62, 86-91,
130-136, 162-182, 211-224, 233-247, 278-288, 1 pi.
1877. Die stammviiter unserer hunde-rassen. Wien. (See also Arch,
f. naturg., 1877, 43, pt. 2, p. 80).
Jentink, F. A.

1897. The dog of the Tengger. Notes Leyden mus., 18, p. 217-220, pi.
4,5.
Keller, Conrad.

1904; Zur abstamniuugsgeschichte unserer hunde-rassen. Eine abwehr
gegeniiber Herrn. Prof. Th. Studer. Vierteljahrsschr. d. Naturf. ges.
Zurich, 48, p. 440-457.
Keller, Otto.

1909. Die antike tierwelt. Leipzig, 8vo, illus.

1912. Die herkunft der europjiischen haustierwelt. Verh. 8 Int. zool.-
kongr., p. 356-365.
Kohler, E. M.

1896. Hunderassen Chinas und der Mongolei. Zool. garten, 37, p.
257-263.
Kohn, F. G.

1911. Beitrag zur kenntnis der haut der nackthundes. Zool. jahrb.
Anat., 31, p. 427-438, pi. 16.
KrefEt, Gerard.

1866. On the vertebrated animals of the Lower Murray and Darling,
their habits, economy, and geographical distribution. Trans. Phil,
soc. N. S. Wales, 1, p. 1-33.
Kreyenberg, .

1910. Chinesische hunderassen und chinesische jagdhunde. Zool.
beobachter (Zool. garten), 51, p. 289-292.

Kiihn, Jxilius.

1887. Fruchtbarkeit der bastarde von schakal und haushund. Zool.
garten, 28, p. 161-163.
Kulagin, Nicolaus.

1892. Mittheilung iiber die hunderasse Laika (Eskimohunde) in Russ-
land. Zool. jahrb. Syst., 6, p. 435-441, pi. 20.
Kunstler, J., and Chaine, J.

1906. Anomalie dentaire du chien. Proces-verb. seances Soc. .sci. phys.
et nat. Bordeaux for 1905-1906, p. 26-27.

Lambe, L. M.

1907. Note on the occurrence of a supernumerary tooth in a dog.
Ottawa nat., 21, p. 25-26, fig.

Landois, Hermann.

1883. Ueber ein anatomisches unterscheidungsmerkmal zwischen haus-
hund und wolf. Morph. jahrb., 9, pt. 1, p. 163-165.
Langkavel, Bernh.

1S92. L'eber Dingos, Pariah-und neuseelandische hunde. Zool. garten,
33, p. 33-38.

510 bulletin: musp:ttm of compakative zoology.

1899. Dogs and savages. Aim. rept. Regents Smithson. inst. for 1898,
p. 651-675 (transl. from Intern, arch, ethnogr., 8, p. 109-149).
Lataste, Fernand,

1888. La premiere premolaire du cliien est-elle monophysaire ou diphy-
saire? Le naturaliste, ser. 2, 10, p. 20-22.

LeConte, John.

1856. Notice of American animals, formerly known, but now forgotten
or lost. Proc. Acad. nat. sci. Phila., 7, p. 8-14.
Lendenfeld, Robert von.

1889. Bilder aus dem australischen urwald. IV. Der Dingo. Zool.
garten, 30, p. 43-49, fig.

Leseble, L.

1890. Les dingos au chenil du Jardin zoologique d'acclimatation. Rev.
sci. nat. appliquees. Bull. mens. soc. nat. d'acclim. de France, 37, p.
681-684.

Lewis, Meriwether, and Clark, William.

1905. Original journals of the Lewis and Clark expedition. 1804r-1806,
etc. . . . edited by Reuben G. Thwaites. 8vo, 7 vols, illus.
Livermore, S. T.

1877. A history of Block Island, [R. I.] from its discovery, in 1514, to
the present time, 1876. Hartford, Conn.
Lonnberg, Einar.

1919. Remarks on some South American Canidae. Arkiv zool., 12,
no. 13, 18 pp., 4 figs.
Looxnis, F. B., and Young, D. B.

1912. On the shell heaps of Maine. Amer. journ. sci., ser. 4, 34, p. 17-
42, figs.
Lord, J. K.

1866. The naturalist in Vancouver Lsland and British Columbia. Lon-
don, Svo, 2 vols., illus.
Lortet, — — , and Gaillard, C.

1909. La favme momifiee de I'ancienne Egypte et recherches anthropo-
logiques. Arch. Mus. d'hist. nat. Lyon, 10, art. 2, 336 pp., 223 figs.
Lucas, F. A.

1897. A dog of the ancient Pueblos. Science, new ser., 5, p. 544.
McCoy, Frederick.

1862. Note on the ancient and recent natural history of Victoria. Ann.
mag. nat. hist., ser. 3, 9, p. 137-150.
MacCurdy, G. G.

1914. The passing of a Connecticut rock shelter. Amer. journ. sci.,
ser. 4, 38, )). 511-.522, pi. 4.
MacFarlane, R.

1905. Notes on mammals collected and observed in the northern
^Mackenzie River district, Northwest Territories of Canada, with
remarks on explorers and explorations of the Far North. Proc. U. S.
N. M., 28, p. 673-764, pi. 30-34.

ALLEN: DOGS OF TH?: AMERICAN ABORIGINES. 511

Manners-Smith, J.

1902. Wolf h5^brids in Gilgit. Journ. Asiatic soc. Bengal, 70, pt. 2,
p. 143-146.
Martinoli, C.

1908. Algunas observaciones sobro las anomalias dentarias del perro.
An. Mus. nac. Buenos Aires, ser. 3, 9, p. 459-469, table.
Mercer, H. C.

1897. An exploration of aboriginal shell heaps revealing traces of canni-
balism on York River, Maine. Publ. Univ. Penna. philol., lit. and
arch., 6, p. 111-137, fig. 40-44.
Miller, G. S., Jr.

1912. Catalogue of the mammals of western Europe (Europe exclusive

of Russia) in the collection of the British museum. London, 8vo, illus.

1912a. The names of the large wolves of northern and western North

America. Smithsonian misc. coll., 59, no. 15, 5 pp.
1916. Bones of mammals from Indian sites in Cuba and Santo Domingo.
Smithsonian misc. coll., 66, no. 12, 10 pp., 1 pi.
Mills, W. C.

1906. Baum prehistoric village. Ohio arch, and hist, publ., 15, p. 44-
136, 85 text-figs.
Mivart, St. George.

1890. Note on canine dental abnormalities. Proc. Zool. soc. London,

p. 376-378, fig.
1890a. Dogs, jackals, wolves, and foxes: A monograph of the Canidae.
London, 8vo, illus.
Mosch, O.

1906. Beitrag zur kenntnis des wolfes. Verh. Schweiz. naturf. ges.
St. Gallen, 89, p. 72-73.
Molina, J. I.

1808. The geographical, natural and civil history of Chile. Middletown,
Ct., 2 vols., 8vo.
Moore, C. B.

1892-3. Certain shell heaps of the St. John's River, Florida, hitherto
unexplored. Amer. nat., 1892, 26, p. 912-922, pi. 24; 1893, 27, p.
8-13, 113-117, 60.5-624, 708-723:
1894. Certain sand mounds of the St. John's River, Florida. Journ.
Acad. nat. sci. Phila., ser. 2, 10, p. .5-103, illus.

1896. Additional mounds of Duval and Clay Counties, Florida. Mound
investigation on the east coast of Horida. Certain Florida coast
mounds north of the St. Johns River. [Philadelphia; privately printed],
4to, 30 pp., 2 ])]s.

1897. Certain aboriginal mounds of the Georgia coast. Journ. Acad,
nat. sci. Phila., ser. 2, 11, p. 5-138, illus.

1899. Certain aboriginal mounds of the coast of South Carolina. Journ.
Acad. nat. sci. Phila., ser. 2, 11, p. 146-166.

512 bulletin: museum of comparative zoology.

1907. Crystal River revisited. Journ. Acad. nat. sci. Phila., .ser. 2, 13,
p. 406-425, illus.
Mortillet, G. de.

1889. Le chien. Bull. 8oc. d'anthrop., ser. 3, 12, p. 425-448.
Morton, S. G.

1850. [On the antiquity of some races of dogs]. Proc. Acad. nat. sci.
Phila., 5, p. 85-89.

1851. [On American Indian dogs.] Proc. Acad. nat. sci. Phila., 5, p.
139-140.

Nathusius, Hermann von.

1874. Bemerkimgen iiber die schadel der Eskimohunde. Die zweite
deutsche Nordpolar fahrt in den jahren 1869 und 1870, 2, p. 175-
177.
Nehring, Alfred.

1882. Ueber einige Canis-schadel mit auffalliger zahnformel. Sitzb.
Ges. naturf. freunde Berlin, no. 5, p. 65-68.

1883. Ueber einige nachtrJiglich an der stiitte des Spandauer Bronze-
fundcs zum vorschein gekommene menschliche und thierische reste.
Zeitschr. ethnol., 15, p. 357-3.59.

1884. Ueber eine grosse wolfsahnliche hunde-rasse der vorzeit {Cards
fain, decumanus Nehring) und ueber ihre abstammung. Sitzb. Ges.
naturf. freunde Berlin, p. 153-165.

lS84a. Ueber Rassebildung bei den Inca-hunden aus den grabern von
Ancon. Kosmos (Zeitschr. ges. entw.-lehr.j, 15, p. 94-111, 3 text-figs.

1884b. Mammals. In Reiss and Stiibel's Peruvian antiquities. The
necropolis of Ancon in Peru, vol. 3, pt. 15, London and BerUn, folio.

1885. I^eber rassenbildung bei den Inca-himden von dem todtenfelde
bei Ancon in Peru. Sitzb. Ges. naturf. freunde Berlin, p. .5-13.

lS85a. Ueber altperuanische hundemumien und ueber rassebildung bei
den sogenannten Inca-hunden. Verh. Berlin, ges. anthrop., ethn. u.
urgcsch., p. 518-521.

1886. Ueber eine neue sendung mumificirter Inca-hunde von Ancon in
Peru. Sitzb. Ges. naturf. freunde Berlin, p. 100-102.

1887. Ueber Ctinn rutilans von Java und Lupus japonicun von Nippon.
Sitz. Ges. naturf. freunde Berlin, p. 66-69.

1887a. Ueber die mumie eines langhaarigen Inca-hundes von Ancon

in Peru. Sitzb. Ges. naturf. freunde Berlin, p. 1.39-141.
1887b. Zur abstammung der hunde-rassen. Zool. jahrb., 3, p. 51-.58.

1890. Ueber Cuon aljniius fotssilis Nehring, nebst bemerkungen liber
einige andere fossile Caniden. Neues jahrb. min., pt. 2, p. 34-52, pi.
2.

Newcombe, C. F.

1909. Guide to the anthropological collection in tlie Provincial Museum.
Victoria, B. C., 8vo, illus.
Noack, Th.

1887. Wolfbastarde. Zool. garten, 28, p. 106-111.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 513

1907. Wolfe, schakale, vorgeschichtliche und neuzeitliche haushunde.
Zool. anz., 31, p. 660-695.

1908. Ueber die abstammung nordrussischer haushunde. Zool. anz.,
33, p. 2.54-264.

1909. Haustiere der Altai-kalmiicken. I. Hunde und katzen. ZooL
anz., 34, p. 683-695.

1910. Ostafrikanische haushunde des Nyassagebietes. Zool. anz., 36,
p. 337-347.

Otto, Hugo.

1909. Beobachtungen an verwilderten hunden. Zool. beobachter
(Zool. garten), 50, p. 176-178.

Oviedo y Valdes, G. F.

1535. Historia general y natural de las Indias, islas y tierra-firme del
Mar Oceano. (Reprint, Madrid, 4 vols, 4to, 1851-55).
Packard, A. S.

1885. Origin of the American varieties of the dog. Amcr. nat., 19,
p. 896-901. Zoologist, 1885, ser. 3, 9, p. 367-372.

Paris, Paul.

1910. Un cas d'hybridation entre loup et chien. Bull. Soc. zool. France,
35, p. 58-59.

Pelzeln, August von.

1886. Eine studieiiber die abstammung der hunderassen. Zool. jahrb.,
1, p. 225-240.

Philippi, R. A.

1886. Ueber die v^eranderungen welche der mensch in der fauna Chile's
bewirkt hat. Festschr. Yer. f. naturk. zu Cassel zur 50-jalu-. beste-
hens, p. 1-20.
Phillips, E. C.

1885. The dog of sacred history. Zoologist, ser. 3, 9, p. 1-3.
Pictet, F. J.

1853. Traite de paleontologie ou histoire naturelle des animaux fossiles
consideres dans leurs rapports zoologiques et geologiques. Paris, ed. 2,
8vo, 4 vols.
Pietrement, C. A.

1888. L'origine et revolution intellectuelle du chien d'arret. Bull.
Soc. anthrop., ser. 3, 11, p. 320-365.
Pocock, R. I.

1900. Hybrid dog and fox. Zoologist, ser. 4, 4, p. 477-478.
Putnam, F. W.

1896. [Remarks on dog-skulls from Ohio]. In Moore, C. B., 1896, p.
26-27.
Rasse, Baron H. de.

1879. Les chiens sauvages et la grande perdrix de La Plata. Bull. Soc.
acchmatation Paris, ser. 3, 6, p. ,389-393.
Recchi, N. A., and Lynceus, J. T.

1651. Rerum medicarum Novae Hispaniae thesaurus sen plantarum,

514 bulletin: museum of comparative zoology,

animalium, mineralium mexicanorum Historia ex Francisci Hernan-
dez .... coUecta, etc. Rome, 4to.
Rengger, J. R,

1830. Naturge.schichte der saugethiere von Paraguay. Basel, 8vo.
Richardson, John.

1829. Fauna Boreali-americana ; or the zoology of the northern parts of
British America. 1, Quadrupeds. London, 4to.
Ross, B. R.

1861. A popular treatise on the fur-bearing animals of the Mackenzie
River district. Can. nat. & geol., 6, p. 5-36.

Ross, J. C.

1835. Account of the objects in the several departments of natural
history, seen and discovered during the present expedition. In Ross,
■John, Appendix to the Narrative of a second voyage in search of a
north-west passage, etc. London, 4tOf illus.
RUtimeyer, L.

1860. Untersuchung der tierreste aus den pfahlbauten der Schweiz.
Mitth. Antiquar. ges. Ziiiich, 13, pt. 2.

1862. Die Fauna der pfahlbauten der Schweiz. Neue denkschr.
allgeni. Schweiz. ges. naturw., 19, p. 1-248, pi. 1-6.

Saint-Loup, Remy.

1892. Les ancetres de nos chiens. Le naturaliste, 14, p. 101-102, 3
text-figs.
Schinz, H. R.

1844. Systematisches verzeichniss aller bis jetzt bekannten saugethiere
Oder Synopsis Mammahum nach dem Cuvier'schen System. Solo-
thurn, 8vo.
Schmerling, P. C.

1833-4. Recherches sur les ossemens fossiles decouverts dans les cavernes
de la Province de Liege. Liege, 2 vols., atlas.
Schmitt, F.

1903. Ueber das postembryonale wachstum des schadels verschiedener
hunderassen. Arch, naturg., 69, pt. 1, no. 1, p. 69-134, pi. 4-5, table.
Seler, E.

1890. [The dog in ancient Mexico]. Congr. Americanistes, BerUn,
1888, sess. 7, p. 321-324.
Smith, C. Hamilton.

1840. The naturalist's library. Mammaha. X. Dogs. Edinburgh,
16 mo., Ulus.
Spegazzini, Carlos.

1882. Costumbres de los habitantes de la Tierra de Fuego. An. Sec.
cient. Argentina, 14, p. 159-181.
[Stockdale, Arthur].

1917. The hairless dog. Journ. heredity, 8, p. 519-520, figs. 21, 22.
Strobel, Pellegrino.

1880. Le razze del cane nelle terramare dell' Emilia. Bull, paleont. Ital.,
6, p. 13-25, 46-53, 110-130, 140-154, 169-179.

ALLKJN": DOGS OF THE AMERICAN ABORIGINES. 515

Strobel, Pellegrino and Pigorini, Luigi.

1864. Le terremare e la ])alafitte del Parmense. Seconda relazione. Atti
Soc. Ital. sci. nat. Milano, 7, art. 2, 152 pp., plate.

Studer, Theophil.

1879. Beitrag zur kenntniss der hvinderacen in den pfahlbauten. Arch,
anthrop., 12, p. 67-78, pi. 2.

1883. Die tliierwelt in den pfahlbauten des Bielersee's. Mitth. Naturf.
ges. Bern, for 1882, pt. 2, Abhandlungen, p. 17-115, pi. 1-5.

1884. Nachtrag zu dem aufsatze iiber die thierwelt in den pfahlbauten
des Bielersee's. Mitth. Naturf. ges. Bern, for 1883, pt. 1, Abhandlungen,
p. 3-26, pi. 1-6.

1901. Die praehistorischen huiide in ihrer beziehung zu den gegenwartig
lebenden rassen. Abh. Hehweiz. paliiontol. ges., 28, no. 1, 137 pp.,
9 pis., 18 text-figs.

1904. Ueber den deutschen schaferhund und einige kynologische fragen.
Mitth^ Naturf. ges. Bern, for 1903, p. 17-55, pi. 1-9.

1905. r^tude sur un nouveau chien preiiistorique de la Russie. L'an-
thropologie, 16, p. 269-285, 2 text-figs.

1906. Ueber einen hund aus der palaoUthischen zeit Russlands. Canis
poutiatini. Zool. anz., 29, p. 24-35, pi. 1-2.

1908. Schadel eines hundes aus einer prahistorischen wohnstatte der
hallstattzeit bei karlstein, amstgericht Reichenhall. Mitth. Naturf. ges.
Bern, for 1907, p. 155-168, 2 pis.

Suckley, George, and Gibbs, George.

1860. Report upon the mammals collected, on the Survey. Rept. Expl.
and surv. railroad from the Mississippi river to the Pacific Ocean, 12,
Washington, 4to.
Teit, James.

1900. The Thompson Indians of British Columbia. Mem. Amer. mus.
nat. hist., 2, p. 163-392, pi. 3.

1909. The Shuswkp. Mem. Amer. mus. nat. hist., 4, p. 443-789, pi.
13-14.

Thorndike, T. W.

1911. The Indian sled dogs of North America. Recreation, February,
p. 74-77, 101-102.
Tiemann, Fr.

1865. Meine hiindin. Bastard von einem nackten afrikanischen hunde
und einer hiesigen wachtelhtindin. Zool. garten, 6, p. 430-434.

Toni, Etorre de.

1897. Sopra un ibrido naturale di Canis familiaris e C. vulpes. Atti R.
ist. Veneto sci. left, ed arti, ser. 7, 8, p. 912-916.
Toussaint, H.

1876. Des rapports qui existent, chez le chien, entre le nombre des dents
molaires et les dimensions des os de la face. Compt. rend. Acad. sci.
Paris, 82, p. 754-756.

516 bulletin: museum of comparative zoology.

Tozzer, A. M., and Allen, G. M.

1910. Animal figures in the Maya codices. Papers Peabody mus., 4,
p. 273-372, pi. 1-39.

Trouessart, £. L.

1911. Le loup de I'lnde {Canis palKpes Sykes), souche ancestrale du
chien domestique. Compt. rend. Acad. sci. Paris, 152, p. 909-913, 2
figs.

Tschudi, J. J. von.

1844-6. Untersuchungen iiber die Fauna Peruana. St. Gallen, 4to,
illus.
Vancouver, George.

1798. A voyage of discovery to the North Pacific Ocean, and round the
world, etc. London, 4to, 3 vols., illus.
Vieira, L.

1894. Etude comparative du squelette du chien et du loup. Annaes sci.
nat., 1, p. 109-114.
Watson, James.

1909. The dog book. New York, large 8vo, illus. ,
Warren, Ernest.

1910. On a black, hairless duiker and dog and a bulklog-headed calf.
Ann. Natal mus., 2, p. 235-252, pi. 10-13, 3 text-figs.

White, Taylor.

1890. On the wild dogs of New Zealand. Trans, and proc. New Zealand

inst., 22, p. 327-330.
1892. On the native dog of New Zealand. Trans, and proc. New Zea-
land inst., 24, p. 540-557.
Wilckens, Martin.

1885. Uebersicht liber die forschungen auf dem gebiete der palaonto-
logie der haustiere. 8. Die hundeartigen tiere (Caniden) des Dilu-
viums. Biol, centralbl., 5, p. 597-604, 621-627.

1886. Uebersicht iiber die forschungen auf dem gebiete der palaontologie
der haustiere. 9. Die vorgeschichtlichen und die pfahlbau-hunde.
Biol, centralbl., 5, p. 719-729, 751-757.

Windle, B. C. A., and Humphreys, J.

1890. On some cranial and dental characters of the domestic dog. Proc.
Zool. soc. London, p. 5-29.
Winship, G. P.

1904. The jomney of Coronado, 1540-1542, from the City of Mexico to
the Grand Canon of the Colorado and the Buffalo Plains of Texas,
Kansas, and Nebraska as told by himseK and his followers. Trans-
lated and edited with an introduction. New York, 12 mo. illus.
Wissler, Clark.

1917. The American Indian. An introduction to the anthropology of
the New World. New York, 8vo, illus.

ALLEN: DOGS OF THE AMERICAN ABORIGINES. 517

Woldrich, J. N.

1877. Ueber einen neiien haushimd der Bronzezeit {Canis familiaris

intermedius) aus den aschenlagen von Weikersdorf, Pulkau und Ploscha.

Mitth. Anthrop. ges. Wien, 7, p. 61-85, pi. 1-5.
1879. Uebei- Caniden aus dem Diluvium. Denkschr. K. akad. wiss.

Wien, math.-nat. cl., 39, pt. 2, p. 97-148, pi. 1-6.
1882. Beitrage zur geschichte des fossilen hundes, nebst beinerkungen

liber die lossbildung. Mitth. Anthrop. ges. Wien, 11, p. 8-17, pi. 1.
1882a. [Die bisherigen resultate der forschung beziiglich des prahistor-

ischen hundes.] Mitth. Anthrop. ges. Wien, 12, p. 27-31.
1882b. Ueber den haushund von Stradonic. Mitth. Anthrop. ges. Wien,

12, p. 153-154.
1886. La descendance des races du chien domestique en Europe.

L'homme, Journ. illus. sci. anthrop., 3, p. 65-68.
1886a. Zur frage iiber die abstammung der europaischen hunderacen.

Anz. K. akad. wiss. Wien, math.-nat. cl., 23, p. 12-16.
1898. Uebersicht der wirbelthierfauna des " Bohmischen Massivs "

wahrend der anthropozoischen epoche. Jahrb. K. k. geol. reichsanstalt

Wien, 47, p. 394-428.
Wolfgramm, Albert.

1894. Die einwirkung der gefangenschaft auf die gestaltung des wolfs-

schadels. Zool. jahrb. Syst., 7, p. 773-822, pi. 24-26.
Wyman, Jeffries.

1868. An account of some kjoekkenmoeddings, or shell-heaps, in Maine

and Massachusetts. Amer. nat., 1, p. 561-584, pi. 14-15.
Yarrell, William.

1833. [On a specimen of the hairless Egj^tian variety of the familiar

dog]. Proc. Zool. soc. London, pt. 1, p. 113.
Zacher, G.

1898. Wie wurde der hund zum haustiere? Natur, 47, p. 397-399.

EXPLANATION OF PLATES.

PLATE 1.

Allen. — Dogs of the American Aborigines.

PLATE 1.

Fig. 1. — Eskimo Dog. The grandparents of this dog were brought by Peary
from Smith's Sound,. Greenland. Photo by Ernest Harold Baynes.

Fig. 2. — The Hare-Indian Dog of northern Mackenzie. From Richardson's
plate (1829).

BULL. MUS.COMP. ZOOL.

Allen. Dogs. Plate 1

PLATE 2.

Allkn . — Dogs of the American Aborigines.

PLATE 2.

Fig. 1. — Mexican Hairless Dog. Reproduction of figure of Lupus vjexicanus
from Recchi and Lynceus (1651).

Fig. 2. — Mexican Hairless Dog, 9 . Photograph by Arthur Stockdale of
Mexico City. Courtesy of The Journal of Heredity.

BULL. MUS.COMP. 200L.

Allen. Dogs. Plate 2

PLATE 3.

Ai,t.EN. — Dogs of (he American Aborigines.

PLATE 3.

Fig. 1. — The Ytzcuinteporzotli or Canis mexicana of Hernandez, reproduced
from the figure by Reechi and Lynceus (1651) . It probably repre-
sents a Raccoon.

Fig 2. — On the right a Mexican Hairless Dog, on the left a hairy dog from
the same litter. The parents of these two were a Mexican Hairless
Dog shown in Plate 2, fig. 2, and a mongrel dog, normally haired.
Courtesy of the Journal of Heredity.

BULL. MUS.COMP.ZOOL.

Allen. Dogs. Plate 3

PLATE 4.

AiXEN'.- — Dogs of the American Aborigines.

PLAICE 4.

Fig. 1. — Clallam-Indian Dog. From the painting by Paul Kane in 1846,
now in the Royal Ontario Museum of Archaeologj' at Toronto.

Fig. 2. — Fuegian Dog. Reproduction of d'Herculais' (1884) figure drawn
from a dog brought to France from Tierra del Fuego by the Mission
Scientifique du Cap Horn.

BULL. MUS. COMP. ZOOL.

Allen. Dogs. Plate 4

PLATE 5.

Allen. — Dogs of llio American Aborigines.

PLATE 5.

Fig. 1. — A dog of the Bersimis Indians, Canada, supposed to represent the
Short-legged Indian Dog. Photograph by William B. Cabot.

Fig. 2.- — Small yellow-and-white or brindle dogs, with a child of the Macusi
Indians in southern British Guiana. These dogs may have more or
less blood of European stock, but probably retain some aboriginal
characteristics. Photograph by Dr. William C. Farrabee.

BULL. MUS. COMP. ZOOL.

Allen. Dogs. Plate 5

PLATE 6.

Allen. — Dogs of the American Aborigines.

PLATE 6.

The Short-nosed Indian Dog (" Pachycyon"). A mummified specimen col-
lected by Messrs. S. J. Guernsey and A. V. Kidder in the Marsh Pass
region, Arizona, and now in the Peabody Museum of Archaeology.
Photograph by S. J. Guernsey.

BULL. MUS. COMP. 200L.

Allen. Dogs. Plate

PLATE 7.

Allen. — Dogs of the American Aborigines.

PLATE 7.

Skull of the Common Indian Dog, collected by Kennicott on Peel River,
northern Mackenzie, U. S. N. M. 6,219. Length 177 mm.

Fig. 1. — Cranium in profile showing relatively weak crests and slender muzzle.

Fig. 2. — Lower ramus; the first premolar normally lacking.

Fig. 3. — Cranium, ventral view; upper first premolar lacking.

BULL. MUS. COMP. ZOOL.

Allen. Dogs. Plate 7

PLATE 8.

Allen. — Dogs of the American Aborigines.

PLATE 8.

Cranium of the Common Indian Dog from Le Moine shell-heap, French-
man's Bay, Maine, collection of Phillips Academy, Andover, Mass., 53,902 Me.
Length 192 mm.

Fig. 1.— Profile view.

Fig. 2. — Ventral view. The first upper premolar is lacking.

BULL. MUS. COMP. ZOOL.

Allen. Dogs. Plate 8

PLATE 9.

Allen. — Dogs of the American Aborigines.

PLATE 9.

Cranium of an luca Dog, collected by Dr. A. Hrdli^ka at Huacho, Peru,
U. S. N. M. 176,309. Length, occiput to anterior root of incisors, 178 mm.

Fig. 1.— Profile.

Fig. 2. — Ventral view. The first premolar is present on the left side only.

BULL. MUS.COMP.ZOOL.

Allen. Dogs. Plate 9

- •>»-

f^U,

PLATE 10.

Alltn. — Dogs of Uie American Aborigines.

PLATE 10.

Small Indian Dog or Techichi, from a cranium collected by L. F. Carr, in
Ely Cave, Lee County, Virginia, M. C. Z. 7,123. Length, occiput to tip of
premaxillaries, 140 mm.

Fig. 1.— Profile.

Fig. 2. — Ventral view.

BULL. MUS. COMP. ZOOL.

Allen. Dogs. Plate 10

PLATE il.

AixBN. — Doga of the American Aboriginest

PLATE 11.

Cranium of a Short-nosed Indian Dog (" Pachycyon ") from shell-mound
on San Nicolas Island, off southern California, Univ. of Cal., Anthrop. Mxis.,
T^W. Length, occiput to tip of premaxillary, 138 mm.

Fig. L— Profile.

Fig. 2. — Ventral view.

BULL. MUS. COMP. ZOOL.

Allen. Dogs. Plate 11

PLATE 12.

Allen. — Dogs of the American Aborigines.

PLATE 12.

Skull of the Peruvian Pug-nosed Dog, collected by Dr. A. Hrdlicka at
Huacho, Peru, U. S. N. M. 176,307. Length of cranium, occiput to tip of
premaxillaries, 147 mm.

Fig. 1. — Profile, showing undershot jaw.

Fig. 2. — Cranium, ventral view.

BULL. MUS. C0MP.200L.

Allen. Dogs. Plate 12

JUO/

ACME

BOOKBINDiNG CO., IMC.

(NOV 2 9 1983

100 CAMnf?I.Or,E ST.-?EET
j CHARLESTOWN. I^iASS. 1

Date Due

•f-Er^^ ^^7

JAN 31 1937
MAR 9--i::£f}00

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Texts

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Bulletin of the Museum of Comparative Zoology at Harvard College"

See other formats